High Radiological Severity in CRSwNP Is Associated With a Mixed Type 2/Type 3 Epithelial Inflammatory Signature

ABSTRACT

Background

Severe chronic rhinosinusitis with nasal polyps (CRSwNP) is associated with a substantial clinical burden and frequently exhibits resistance to standard therapies. However, the precise inflammatory profiles driving disease severity remain incompletely understood, limiting the development of targeted interventions.

Methods

We prospectively recruited 48 participants, including 38 patients with CRSwNP. Patients were stratified based on their preoperative Lund–Mackay Score (LMS) into low (LMS ≤ 12, n = 27) and high (LMS > 12, n = 11) severity groups; 10 healthy controls were also enrolled. Clinical data were analyzed alongside histopathological characteristics of nasal polyp tissues. Additionally, sinus mucosal brushings underwent RT‐qPCR to profile inflammatory gene expression.

Results

Patients in the high LMS group demonstrated a significantly greater clinical burden, as indicated by endoscopic scores, nasal symptom scores, and peripheral eosinophil counts. Histological analyses revealed that high LMS correlated with extensive tissue remodeling, including basement membrane thickening, goblet cell hyperplasia, and pronounced mixed eosinophilic and neutrophilic infiltration. Molecular profiling identified a distinct mixed inflammatory signature in the high LMS group: in addition to the expected elevation of Type 2 markers (e.g., IL‐5, IL‐13, and CCL26), Type 3 inflammatory genes (e.g., CXCL2, IL‐1β, and IL‐8) were also significantly upregulated.

Conclusion

Severe CRSwNP, characterized by a high LMS, exhibits a mixed Type 2/Type 3 epithelial inflammatory signature. The co‐activation of Type 3 pathways alongside Type 2 responses is associated with severe tissue remodeling. These findings further elucidate the complex inflammatory landscape of CRSwNP and suggest that high radiological scores may reflect distinct molecular heterogeneity in patients exhibiting extensive sinus opacification.

This study demonstrates that high radiological severity in patients with CRSwNP is uniquely characterized by a mixed Type 2 and Type 3 epithelial inflammatory signature. These findings suggest that severe disease involves a complex immune endotype, providing new insights into precision therapeutic strategies for refractory CRSwNP.

1. Introduction

Severe chronic rhinosinusitis with nasal polyps (CRSwNP) is a heterogeneous inflammatory disorder characterized by persistent mucosal inflammation and polyp formation, imposing a substantial burden on patient quality of life and healthcare systems [1]. Type 2 inflammation, driven by cytokines such as IL‐4, IL‐5, and IL‐13, is recognized as the predominant inflammatory endotype in Western populations [2, 3]. However, CRSwNP presents a complex inflammatory landscape. Type 2 inflammation is typically characterized by eosinophilic infiltration, elevated serum IgE, and epithelial hyperplasia [4, 5, 6]. Nevertheless, a substantial subset of patients, particularly those with severe and persistent disease, demonstrates suboptimal responses to corticosteroids and novel Type 2‐targeted biologics. This therapeutic resistance suggests involvement of additional inflammatory pathways in disease severity among refractory patients.

The Lund–Mackay Score (LMS), derived from computed tomography (CT), is a standard quantitative metric for assessing sinus opacification and disease burden [7, 8, 9, 10]. Beyond evaluating the extent of anatomical involvement, the LMS has also been recognized as a potential prognostic indicator. Recent studies have demonstrated that in eosinophilic CRSwNP, higher LMS scores correlate with more extensive disease, increased recurrence risk, and worse postoperative outcomes [7, 11]. Furthermore, elevated LMS has been associated with increased blood and tissue eosinophil levels [12].

Despite these associations, the precise immunological mechanisms underlying severe forms of CRSwNP, characterized by high LMS, remain incompletely understood. Traditionally, research has focused on a binary classification of inflammation as eosinophilic or non‐eosinophilic. This dichotomy may oversimplify the pathological complexity, potentially neglecting intricate mixed inflammatory pathways. Although high LMS correlates with eosinophilia, it remains unclear whether radiological severity results solely from intensified Type 2 inflammation or if other inflammatory mechanisms are recruited. The role of Type 3 inflammation in CRSwNP has gained increasing attention recently. Type 3 inflammation is defined by IL‐17‐mediated pathways and neutrophil‐associated immune responses. Several studies have reported elevated expression levels of IL‐17A, IL‐22, and neutrophil‐related mediators in subsets of patients with CRSwNP [13, 14, 15].

We hypothesized that severe CRSwNP, defined by a high LMS, reflects not only amplified Type 2 responses but also a mixed inflammatory pattern involving both Type 2‐ and Type 3‐mediated pathways. To test this, we performed a comprehensive characterization of LMS‐stratified CRSwNP patients by integrating clinical features, histopathological data, and molecular inflammatory profiles. Identifying a mixed Type 2/Type 3 signature in severe disease could provide mechanistic insights into treatment resistance and inform novel therapeutic targets.

2. Materials and Methods

2.1. Patients and Study Design

This was a single‐center prospective study that recruited 48 adult patients (age ≥ 18 years) undergoing surgical treatment at the Department of Otolaryngology, Seventh Affiliated Hospital of Sun Yat‐sen University, from February 2024 to February 2025. Of these patients, 38 were diagnosed with CRSwNP based on the European Position Paper on CRS (EPOS) guidelines [16]. The remaining 10 patients were diagnosed with a deviated nasal septum and confirmed to be free of sinusitis. Participants were divided into two groups: an experimental group (38 CRSwNP patients) and a healthy control group (10 patients with a deviated nasal septum). CT scans were evaluated using the LMS system (range: 0–24). Based on previous studies, cut‐off values of approximately 12–13 have been employed to distinguish patients with higher disease and inflammatory burden [11, 17]. Therefore, the 38 CRSwNP patients were further stratified into a low LMS group (0 < LMS ≤ 12; n = 27) and a high LMS group (LMS > 12; n = 11). Exclusion criteria included the use of corticosteroids or other hormonal therapies within 1 month prior to surgery. Additionally, CRSwNP patients with fungal sinusitis, primary ciliary dyskinesia, or cystic fibrosis were excluded. The study protocol was approved by the Ethics Committee of the Seventh Affiliated Hospital of Sun Yat‐sen University (Approval No. KY2024‐134‐01). All participants provided written informed consent prior to data collection.

2.2. Clinical Data Collection

All patient information, including basic demographic data and medical history, was collected upon admission. Patients underwent CT scanning and nasal endoscopy. Two senior radiologists independently evaluated the CT scans using the LMS system, and the mean scores were recorded [18]. Nasal endoscopic findings were assessed by trained endoscopists using the modified Lund–Kennedy (MLK) scoring system (range: 0–12), considering mucosal edema, nasal discharge, and polyps [19]. Patients also completed validated questionnaires: the Sino‐Nasal Outcome Test‐22 (SNOT‐22) and the Tess nasal symptom questionnaire. Peripheral blood samples were drawn preoperatively for complete blood count analysis, including differential leukocyte counts, with both percentages and absolute counts of eosinophils and neutrophils recorded.

2.3. RNA Extraction and Real‐Time Quantitative PCR

Nasal brushing, a non‐invasive sampling method, primarily enriches epithelial cells while also capturing adherent inflammatory cells and immune cell‐derived transcripts on the mucosal surface, as described previously [20, 21]. Nasal epithelial brushing was performed during surgery in the middle nasal meatus of each patient for subsequent gene expression analysis. Total RNA extraction from epithelial samples was carried out using TRIzol Reagent (Takara, Japan), following the manufacturer's protocol. Complementary DNA (cDNA) synthesis was conducted using the PrimeScript RT Master Mix (Takara, Japan). For quantitative real‐time PCR, 50 ng of cDNA from each sample was amplified using the TB Green Premix Ex Taq II kit (Takara, Japan) on a QuantStudio 5 Real‐Time PCR System (Thermo Fisher Scientific, Waltham, MA). Gene‐specific primers are listed in Table S1. 18S rRNA served as an internal normalization control, and relative gene expression levels were calculated via the 2^−ΔΔCt method. Inflammatory markers selected for exploratory profiling were based on previously validated transcriptomic and tissue‐based studies of CRSwNP. Representative cytokines and chemokines associated with Type 1 (e.g., IFN‐γ, CXCL9–11), Type 2 (e.g., IL‐5, IL‐13, CCL26, TSLP), and Type 3 (e.g., IL‐17A, IL‐8, CXCL2, IL‐1β) pathways were analyzed.

2.4. Histopathological Investigations

Nasal polyp tissues were obtained from 38 CRSwNP patients in the experimental group, while concha bullosa tissues were collected from patients in the healthy control group (individuals with deviated nasal septum). Tissues were processed and stained with hematoxylin and eosin (H&E). Paraffin‐embedded sections (4 μm thickness) were deparaffinized in xylene, rehydrated through graded ethanol (100%, 95%, 85%, and 75%), and then rinsed in distilled water. Sections were stained with hematoxylin for 3–5 min, rinsed in running water, and differentiated in 1% acid alcohol when necessary. After bluing in tap water, the sections were counterstained with eosin for 1–2 min, dehydrated in ascending ethanol concentrations, cleared in xylene, and mounted using resin‐based mounting medium with coverslips. Pathological assessment adhered to established standards for nasal epithelium evaluation [22, 23], including basement membrane thickness, epithelial cell counts, and inflammatory cell infiltration (eosinophils and neutrophils). Data were obtained from 8 to 10 randomly selected high‐power fields (HPFs) per section. Goblet cells, eosinophils, and neutrophils were quantified at 400× magnification, averaged across fields, and expressed as mean cell counts per HPF.

2.5. Statistical Analyses

Statistical analyses were conducted using SPSS version 26.0 (IBM Corp., Armonk, NY, USA). Continuous variables with normal distributions were analyzed using one‐way ANOVA for comparisons among three groups, while non‐normally distributed data were analyzed using the Kruskal–Wallis H‐test. Bonferroni‐corrected post hoc tests were applied for pairwise comparisons. Multivariate logistic regression was performed to assess the independence of observed significant factors from potential clinical confounders. Categorical variables were compared using the chi‐square test or Fisher's exact test when expected frequencies were below 5. Correlations were evaluated using Pearson's or Spearman's coefficients based on data distribution. Normally distributed data are presented as mean ± standard deviation (SD), and non‐normally distributed data as median with interquartile range (IQR). Statistical significance was defined as a two‐sided p value < 0.05.

3. Results

3.1. Baseline Clinical and Demographic Characteristics

The study cohort comprised 48 subjects, stratified into three groups: healthy controls (n = 10), low LMS group (LMS ≤ 12; n = 27), and high LMS group (LMS > 12; n = 11). Representative coronal sinus CT images illustrating the radiologic characteristics of each group are shown in Figure 1. Demographic characteristics, clinical symptoms, and laboratory parameters were comparable across groups (Table 1). The high LMS group demonstrated significantly worse clinical outcomes, characterized by severe hyposmia (p = 0.006) and higher MLK scores (p = 0.001). Laboratory analysis indicated significantly elevated peripheral blood eosinophil counts (p = 0.019) and percentages (p = 0.047) in the high LMS group. Bonferroni‐corrected post hoc analysis revealed significantly lower hyposmia scores in the high LMS group compared to healthy controls (p = 0.006) and the low LMS group (p = 0.043). Peripheral eosinophil counts were significantly higher in the high LMS group compared to healthy controls (p = 0.040), though no significant difference was found between the high and low LMS groups. MLK scores were significantly elevated in the high LMS group compared to both healthy controls (p = 0.001) and the low LMS group (p = 0.005). No significant differences in peripheral neutrophil counts were observed among the groups.

FIGURE 1.

Representative coronal sinus CT images from CRSwNP patients with varying LMS. (A, B) Coronal sinus CT images from healthy controls (score = 0); (C, D) Coronal sinus CT images from patients with low LMS (0 < score ≤ 12); (E, F) Coronal sinus CT images from patients with high LMS (score > 12).

TABLE 1.

Demographics and clinical characteristics of the study population.

Variable	Control (n = 10)	Low LMS (n = 27)	High LMS (n = 11)	p
Demographics
Age, years	35.60 ± 13.83	36.78 ± 13.80	37.09 ± 11.07	0.962
Male, n (%)	8 (80.0%)	21 (77.8%)	5 (45.5%)	0.119
Previous nasal surgery, n (%)	0 (0%)	1 (3.7%)	0 (0%)	1.000
Smoking status, n (%)	0 (0%)	7 (25.9%)	1 (9.1%)	0.214
Clinical symptoms
Allergic rhinitis, n (%)	5 (50.0%)	8 (29.6%)	1 (9.1%)	0.119
Asthma, n (%)	0 (0.0%)	0 (0.0%)	2 (18.2%)	0.089
Nasal congestion TNSS	2.0 (1.25, 2.0)	2.0 (1.0, 2.0)	2.0 (1.0, 2.25)	0.628
Runny nose TNSS	1.0 (0, 2.0)	1.0 (1.0, 2.0)	1.5 (1.0, 2.0)	0.493
Postnasal discharge TNSS	0 (0, 1.0)	1.0 (0, 1.0)	1.0 (0, 2.0)	0.313
Hyposmia TNSS	0 (0, 1.0)	1.0 (0, 1.0)	3.0 (0, 2.0)	0.006*
TNSS scores	2 (2, 5)	4 (2, 6.25)	7 (6, 8)	0.063
Modified Lund–Kennedy score	4 (2, 4)	4 (3, 5.25)	7 (6, 12)	0.001***
SNOT‐22 score	17.29 ± 12.19	28.56 ± 14.70	41.42 ± 23.20	0.278
Laboratory findings
Blood eosinophil count (×109/L)	0.13 (0.08, 0.20)	0.17 (0.10, 0.27)	0.29 (0.21, 0.44)	0.019*
Blood eosinophil percentage (%)	2.0 (1.5, 3.0)	2.5 (1.53, 5.03)	4.8 (2.95, 7.48)	0.047*
Blood neutrophil count (×109/L)	3.49 (2.71, 4.29)	3.60 (3.19, 5.26)	3.82 (2.57, 4.98)	0.944
Blood neutrophil percentage (%)	59.80 (51.05, 60.45)	58.95 (52.65, 63.50)	56.85 (55.00, 62.68)	0.637

Note: Data were presented as mean ± standard deviation (SD) for normally distributed variables, and as median with interquartile range (IQR) for non‐normally distributed variables. Categorical variables were shown as number (%). p values were calculated using the one‐way analysis of variance (ANOVA) or Kruskal–Wallis H‐test for continuous variables and the chi‐square or Fisher's exact test for categorical variables, as appropriate. p < 0.05 was considered statistically significant.

Abbreviations: SNOT‐22, 22‐item Sino‐Nasal Outcome Test; TNSS, total nasal symptom score.

^* p < 0.05.

^*** p < 0.001.

3.2. The LMS Correlates With Clinical Symptom Severity

The association between LMS and clinical parameters in CRSwNP patients was assessed. Spearman correlation analysis indicated a significant positive correlation between higher LMS and hyposmia, MLK scores, total nasal symptom scores (TNSS), and peripheral eosinophil counts (Figure 2A,B). Conversely, LMS was not significantly correlated with postnasal drip scores (PNDS), comorbid asthma, or peripheral neutrophil counts. These results suggest that elevated LMS reflects increased local endoscopic severity and systemic eosinophilic inflammation.

FIGURE 2.

Correlation matrix of LMS and clinical parameters in CRSwNP patients. (A) Heatmap illustrating Spearman correlations between LMS and clinical variables: Postnasal drip score (PNDS), hyposmia, asthma comorbidity, modified Lund–Kennedy score (MLK), peripheral eosinophil count (EO), eosinophil percentage (EO%), total nasal symptom score (TNSS), peripheral neutrophil count (Neu), and neutrophil percentage (Neu%). Positive correlations are indicated in red, and negative correlations are in blue. (B) Scatter plots demonstrating significant positive correlations between LMS and hyposmia, MLK, TNSS, and EO.

3.3. Severe CRSwNP Is Characterized by Concurrent Activation of Type 2 and Type 3 Pathways

Inflammatory profiles were characterized by examining the expression of Type 1, Type 2, and Type 3 immune‐related genes in nasal mucosal tissues. Type 1 inflammatory markers showed no significant differences among groups (Figure 3A). Key Type 2 inflammatory genes, such as CLC, CST1, CCL26, IL‐13, IL‐5, and TSLP, were significantly elevated in the high LMS group compared to the low LMS group and healthy controls (Figure 3B).

FIGURE 3.

Differential expression of inflammatory cytokines in nasal polyp tissues. (A) RT‐qPCR analysis of mRNA levels for Th1‐associated cytokines (INF‐G, CXCL9, CXCL10, CXCL11, GZMH, and ZNF683). (B) RT‐qPCR analysis of mRNA levels for Th2‐associated cytokines (CLC, CST1, CCL26, IL‐13, IL‐5, and TSLP). (C) RT‐qPCR analysis of mRNA levels for Type 3‐associated cytokines (CXCL2, IL‐1β, IL‐8, IL‐17A, NOX1, and CCL20). Data are shown as mean ± SD. *p < 0.05; **p < 0.01; ***p < 0.001.

Additionally, Type 3 inflammatory genes (CXCL2, IL‐1β, IL‐8, NOX1, and CCL20) were significantly upregulated in the high LMS group (Figure 3C). Expression of IL‐17A was significantly higher in the high LMS group compared to the low LMS group but did not differ significantly from healthy controls. These findings demonstrate that severe CRSwNP (high LMS) involves a mixed inflammatory profile characterized by co‐activation of Type 2 and Type 3 pathways.

A multivariate logistic regression model was constructed to control for potential clinical confounders, including asthma, allergy, smoking, and surgical history. Results indicated that CCL20 (OR = 3.34, 95% CI: 1.13–9.85, p = 0.029), CLC (OR = 2.67, 95% CI: 1.08–6.62, p = 0.034), and CXCL‐2 (OR = 1.92, 95% CI: 1.04–3.55, p = 0.038) were significantly and independently associated with the high LMS phenotype. Other markers such as CST‐1 (p = 0.063), IL‐1β (p = 0.066), IL‐13 (p = 0.080), and TSLP (p = 0.085) showed a positive correlation trend but did not reach statistical significance (Table S2).

3.4. Histological Analysis Reveals Distinct Tissue Remodeling and Inflammatory Infiltration in Severe Disease

Histological examination of nasal tissues revealed progressive changes corresponding to disease severity (Figure 4A). Healthy controls exhibited a relatively intact epithelial layer with minimal inflammation. The low LMS group showed moderate inflammatory infiltration and tissue remodeling. Patients in the high LMS group displayed the most severe histopathological alterations. Quantitative analysis confirmed significantly increased basement membrane thickness, goblet cell hyperplasia, and counts of eosinophils and neutrophils in the high LMS group compared to the healthy control and low LMS groups (Figure 4B).

FIGURE 4.

Histological comparison of nasal polyp epithelia. (A) Representative high‐magnification H&E‐stained images of nasal polyp tissues from healthy controls, low LMS, and high LMS groups. Images show characteristic inflammatory infiltration and epithelial alterations. Red arrows indicate goblet cell proliferation within respiratory epithelium, identified by swollen, clear cytoplasm. Black arrows indicate eosinophils in the lamina propria, characterized by bilobed nuclei and bright eosinophilic (red) granules. Green arrows indicate neutrophils, characterized by multilobed nuclei and pale‐stained cytoplasm. Scale bar = 50 μm. (B) Quantitative analysis of eosinophil and neutrophil counts, basement membrane (BM) thickness, and goblet cell counts in H&E‐stained sections. Data are presented as mean ± SD. *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001.

4. Discussion

In this study, we observed that CRSwNP patients with high preoperative LMS scores exhibited more severe disease, indicated by worse endoscopic findings, more severe nasal symptoms, and elevated peripheral blood eosinophil counts. Olfactory dysfunction and eosinophilic inflammation are hallmark characteristics of Type 2 chronic rhinosinusitis with nasal polyps (CRSwNP), according to clinical experience [10]. In contrast, isolated Type 3 inflammation typically does not present with olfactory loss or eosinophilia. However, mixed inflammatory patterns (combining Type 2 and Type 3) can still show significant olfactory impairment and eosinophilia. This suggests that features typical of Type 2 inflammation do not exclude concurrent Type 3 inflammation [24]. Thus, high LMS scores in CRSwNP patients represent a complex and heterogeneous disease state, involving multiple overlapping inflammatory pathways contributing to severe and treatment‐resistant phenotypes.

At the molecular level, high LMS scores correlated with increased expression of classic Type 2 inflammatory markers (e.g., IL‐5, IL‐13, and CCL20) and significant upregulation of Type 3 inflammatory mediators (e.g., CXCL2, IL‐1β, and IL‐8), forming a mixed inflammation profile. Multivariate logistic regression identified CCL20, CLC, and CXCL‐2 as significantly associated with the high LMS group. The CCL family, including CCL11, CCL20, and CCL26, plays a crucial role in recruiting eosinophils in Type 2 CRS by binding to the CCR3 receptor on eosinophils. These chemokines also facilitate dendritic cell (DC) infiltration, leading to the differentiation of naïve CD4⁺ T cells into Th2 cells [25]. Charcot–Leyden crystal (CLC) protein, a marker of Type 2 immune responses, strongly indicates severe eosinophilic infiltration and activation in sinus mucosa [26]. The CXCL family (e.g., CXCL‐1, CXCL‐2, CXCL‐5, and CXCL‐8) serves as potent neutrophil chemoattractants in non‐Type 2 inflammation, activating neutrophils via specific CXCR ligands [27, 28]. Additionally, expression levels of IL‐17A and IL‐8 were notably higher in the high LMS group, although they did not achieve significance in the multivariate analysis. IL‐17A promotes neutrophil recruitment and activation by inducing CXCL2 [29, 30, 31], while IL‐8 contributes to mucosal edema [32, 33, 34, 35]. Collectively, these findings support the hypothesis that the high‐LMS phenotype is driven by a mixed inflammatory pattern. However, several typical cytokines, such as IL‐5, IL‐13, and TSLP, were not statistically significant in our multivariate regression, likely due to the small sample size of the high‐LMS subgroup (n = 11), limiting the statistical power to detect independent effects of these cytokines.

From a histological perspective, tissue remodeling was another notable feature in the high LMS group. Basement membrane thickening and goblet cell hyperplasia are characteristic of chronic epithelial injury and persistent inflammation [20, 21, 36]. Additionally, the high LMS group exhibited more pronounced eosinophilic and neutrophilic infiltration. Previous research indicates that increased infiltration of eosinophils and neutrophils correlates with greater radiological disease burden and more severe sinus opacification [37]. These structural and cellular alterations have been associated with polyp persistence, recurrence, and resistance to conventional therapies [38, 39, 40, 41]. However, the cross‐sectional nature of our study prevents establishing causality between these alterations and disease prognosis.

This exploratory study has several limitations. These include a single‐center design and small sample size, particularly in the high LMS subgroup, potentially limiting the generalizability of our findings. Future studies should collect more extensive data from the high LMS group to validate cytological and pathological classifications. Secondly, the sampling method used remains controversial. Nasal brushing primarily reflects superficial mucosal inflammation rather than deeper submucosal tissue inflammation. However, previous studies have shown that transcriptome analyses of brush samples can capture significant features of deeper tissue inflammation, particularly Type 2 immune activation and inflammatory cell infiltration (e.g., ILC2s, mast cells, T cells) [21, 36, 42]. Therefore, nasal brushing remains valuable for classifying sinusitis, offering advantages such as minimal invasiveness, excellent patient tolerability, and suitability for repeated outpatient assessments. Thirdly, this exploratory investigation examined associations among clinical phenotypes, radiologic findings, and cytokine profiles but lacked deeper mechanistic insights into underlying biological processes. Future research should focus on exploring underlying mechanisms through functional experiments at the molecular level and detailed pathological analyses.

5. Conclusion

This small exploratory study highlights the potential utility of LMS in characterizing CRSwNP and identifies a mixed Type 2/Type 3 inflammatory profile in severe cases. These findings expand our understanding of the complex inflammatory landscape and suggest that high radiological scores might indicate distinct molecular heterogeneity in patients with extensive sinus opacification. A more comprehensive approach should be adopted when evaluating CRS patients with high LMS.

Funding

This work was supported by grants from Shenzhen Fundamental Research Program (JCYJ20240813150440052) and Guangdong Basic and Applied Basic Research Foundation (2023A1515111089).

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Table S1: Primer sequences.

Table S2: Multivariate logistic regression analysis for related genes of high‐LMS group in CRSwNP patients.

Data Availability Statement

The data that supports the findings of this study are available in the Supporting Information of this article.