Treatment Approach to Chronic Rhinosinusitis Based on Endotype: Difference Between East and West

Heterogeneity among CRSwNP patients in Western and Eastern countries leads to differences in management. Type 2 (T2) inflammation accounts for the majority of CRSwNP patients in Western countries, while non-T2 inflammation is the main endotype in CRSwNP patients in Eastern countries. Precise identification of T2 inflammation holds significant promise for optimising treatment outcomes. The treatment of T2 inflammation-biased CRSwNP primarily involves glucocorticoids, reboot surgery, and T2 biologics, while non-T2 inflammation-dominant CRSwNP is mainly treated with macrolide antibiotics. Chronic Rhinosinusitis (CRS) is a chronic nasal and paranasal sinus inflammation lasting over 12 weeks, driven by environmental and host factors. It affects 5%–12% of the population, varying by region and ethnicity. CRS significantly impairs quality of life, increases depression risk, and burdens society through healthcare costs and productivity loss. Additionally, both Eastern and Western CRS patients face a significant challenge of recurrence, even after undergoing standard and comprehensive therapeutic strategies [1]. CRS is divided into two phenotypes: CRS without nasal polyps (CRSsNP) and CRS with nasal polyps (CRSwNP). CRSwNP is a heterogeneous disease with varying inflammatory profiles across regions. In Western countries, 80% of CRSwNP patients exhibit type 2 (T2) inflammation. In contrast, among Eastern populations such as those in China, Japan, Korea, and Malaysia, T2 inflammation is observed in only 20%–60% of cases, with neutrophilic inflammation demonstrating higher prevalence [2]. The cytokine patterns also demonstrate distinct variations. In Western CRSwNP patients, 73%–87% exhibit T2 inflammation, predominantly characterised by single IL-5+T2 inflammation. In Asian populations, the prevalence of IL-5+T2 inflammation ranges from 20% to 61%, frequently coexisting with both IFN-γ+T1 and IL-17+T3 inflammation [3]. Severe CRSwNP not only manifests as persistent T2 inflammation, but is also frequently associated with comorbidities including allergic rhinitis (AR), asthma, and aspirin-exacerbated respiratory disease (AERD). AR and asthma are more common in Western CRSwNP patients (AR: 50%–75%, asthma: > 50%) than in Eastern patients (AR: 13%, asthma: 9%). Similarly, AERD prevalence is higher in Western populations (9.69%–35%) than in South China (0.57%), possibly due to diagnosis based on clinical history rather than oral aspirin challenge [2]. Although numerous biomarkers have been identified for characterising endotypes, challenges persist in their application to routine clinical practice: (1) There is a lack of unanimous criteria and consensus regarding a cut-off point for biomarkers identifying T2 inflammation and neutrophilic inflammation. (2) Most biomarkers are identified through laboratory research, but non-invasive biomarkers with good clinical accessibility and simplicity are needed in real-world practice. (3) AI-driven endotyping classification faces limited clinical application due to validation gaps and infrastructural constraints, especially in resource-limited regions. Therefore, the practical strategies to identify T2 inflammation were summarised: (1) Clinical symptoms: hyposmia; (2) Comorbidities: asthma, AERD, AR; (3) CT scan: an opacity with a central lesion; (4) Laboratory tests: serum IgE ≥ 100 IU/mL and blood eosinophil counts ≥ 150/μL; Tissue eosinophil > 10% of inflammatory cells or > 55/high-power field (HPF) [4-6]. Differences in endotypes underscore the necessity for region-specific therapeutic strategies. Corticosteroids are pivotal in treating CRS due to their potent anti-inflammatory effects. Both topical and systemic corticosteroids are currently indispensable for the treatment of CRS and postoperative management of endoscopic sinus surgery (ESS) in both Eastern and Western guidelines [1]. However, their efficacy varies between Eastern and Western patients. Western CRSwNP patients predominantly exhibit T2 inflammation and respond well to corticosteroids, with clinical trials showing ≥ 70% improvement in symptoms and objective outcomes [7]. In contrast, Asian CRSwNP patients, who often have neutrophilic or T3-skewed inflammation, show lower response rates to corticosteroids. Studies in China reported that only 53.9% of CRSwNP patients responded to oral corticosteroids, and combined oral and topical corticosteroids did not significantly enhance efficacy [8]. The advent of steroid-eluting stents, nebulized corticosteroids, and steroid-impregnated packing materials has significantly improved corticosteroid delivery efficiency. Similarly, the therapeutic effects of these devices have mainly been observed in Western populations, whereas there are inconsistent reports regarding their efficacy in Asian populations [1]. This discrepancy is attributed to the primary action of corticosteroids on T2 inflammation, which is less prevalent in Eastern populations. Biomarkers associated with T2 inflammation, including eosinophil counts, IgE levels, and elevated IL-5, serve as important predictors of corticosteroid efficacy, highlighting the crucial need to explore non-invasive detection methods for these biomarkers to optimise corticosteroid therapy in Asia. Although evidence is limited, macrolide antibiotics (e.g., clarithromycin, azithromycin) are used in long-term, low-dose regimens for CRS, and the therapeutic mechanisms are associated with epithelial barrier restoration and reduced T1 and T3 inflammation. Studies have shown that macrolide might be effective for CRS patients with elevated levels of IL-8, IL-6, and IgG4, suggesting that patients with neutrophilic inflammation who respond poorly to glucocorticoid treatment might benefit from macrolides. Of note, macrolides exhibit an add-on effect to oral and nasal steroid in Eastern CRSwNP patients. In contrast, studies in the West generally show no significant benefit of macrolides over placebo, possibly due to the higher prevalence of T2 inflammation in these populations. Therefore, long-term, low-dose macrolide therapy is recommended in Chinese guidelines for treating corticosteroid-resistant CRSwNP patients, neutrophil-dominant nasal polyps, the persistent edema of nasal mucosa, and purulent nasal discharge [9], whereas EPOS2020 provides no clear recommendations due to a lack of high-quality evidence and potential side effects [1]. In summary, macrolide antibiotics may primarily target neutrophilic inflammation and T1- and T3-skewed inflammation, which possibly explains their greater therapeutic efficacy in Asian populations. The levels of IL-8 and IgG4 are considered to be predictive biomarkers for macrolide efficacy. For CRS patients unresponsive to comprehensive medical therapy, endoscopic sinus surgery (ESS) remains the primary intervention to restore sinus ventilation and drainage. With a deeper understanding of the anatomy, aetiology, and pathogenesis, surgical approaches have evolved, ranging from less extensive functional ESS to extended/extensive ESS. Although large-scale RCTs comparing different surgical methods are limited, all types of sinus surgery are performed to alleviate symptoms, improve quality of life, and reduce recurrence [1]. The scope of procedure and indications of various surgeries are provided in Tables S1. With the acceptance of mucosal endotypes, researchers from Ghent University have proposed a new surgical technique—Reboot surgery—as an alternative treatment for the severe T2-biased nasal polyps. The Reboot approach could reduce inflammatory load and reform a functional ciliated mucosal layer in the sinuses. The follow-up data from Europe demonstrated that compared with the conventional ESS approach, Reboot surgery significantly improved olfaction, reduced recurrence rates (from 40%–50% to < 15%), and decreased oral steroid dependency. However, a multicenter study from South China reported that nasalization performed better than partial Reboot in eosinophilic CRSwNP patients after more than 12 months of follow-up, highlighting differences in therapeutic efficacy between Eastern and Western populations. Due to discrepancies in the efficacy of the Reboot procedure between Eastern and Western populations, as well as the limitations of the current studies with small sample sizes, large-scale studies with longer follow-up periods across diverse regions are needed to evaluate the long-term efficacy and safety of this procedure. Nowadays, there is no rigorous evidence-based medicine to determine how the surgical procedures should be performed according to mucosal endotype. However, due to differences in immunological endotype, the old principle "one fits all" is not valid anymore in either the West or the East. For patients with non-T2 CRSwNP, possibly limited to a single or few sinuses and at low risk of recurrence, appropriate management should be conducted according to the principles of Messerklinger and Stammberger. However, CRSwNP patients with a severe T2 immune reaction occupying all sinuses will be managed by the more extensive approaches. Although Reboot surgery is recommended for those patients in Western countries, there is insufficient evidence to support its efficacy in Asian patients. Therefore, whether Reboot surgery should be recommended for CRSwNP patients in Eastern countries needs further research. Increasing clinical trials have witnessed a tremendous boost in the development of monoclonal antibody therapies for the treatment and management of adult severe uncontrolled CRSwNP. Representative biologics targeting T2 and non-T2 inflammatory pathways are illustrated in Figure 1. Phase 3 trials of biologics including dupilumab (anti-IL-4Rα), omalizumab (anti-IgE), mepolizumab (anti-IL-5), benralizumab (anti-IL-5Rα), and tezepelumab (anti-TSLP), were conducted on refractory CRSwNP patients primarily in Western countries, and the clinical data provided convincing proof of good efficacy and safety. While these approved biologics improve patient outcomes, they are required to be administered frequently. Depemokimab, an ultra-long-acting biologic, reduces dosing frequency, potentially enhancing adherence and reducing treatment burden. Representative studies of relevant biologics are summarised in Table S2. The higher prevalence of the T2 endotype in Western populations leads to more frequent use and possibly superior efficacy of T2-targeted biologics in Western patients with CRSwNP compared to Eastern counterparts. However, there is a higher prevalence of non-T2 inflammation in Asian patients, who may exhibit limited responses to these biologic agents. Therefore, it is crucial to first identify appropriate candidates through endotypes. Compared to CRSwNP patients in Western countries, Asian patients exhibit a higher prevalence of neutrophilic and T1/T3-biased inflammation. While biologics such as brodalumab (anti-IL-17RA), secukinumab (anti-IL-17), avacopan (anti-C5a receptor), and emapalumab (anti-IFN-γ) are approved for ankylosing spondylitis, moderate to severe psoriasis, ANCA-associated vasculitis, and hemophagocytic lymphohistiocytosis [2], the effects of these non-T2 biologics on CRSwNP remain underexplored. Interestingly, a case report demonstrated that secukinumab is superior to dupilumab in improving clinical outcomes in CRSwNP patients with non-T2 and mixed endotype, suggesting that IL-17 could be a promising therapeutic target for non-T2 CRSwNP. Therefore, multi-center clinical trials with large-scale non-T2 patients are required to validate these findings. Artificial intelligence (AI) and machine learning are driving innovation in precision medicine for CRSwNP, demonstrating potential in diagnosis and endotype classification across imaging, pathological analysis, and biomarkers in nasal secretions. Although AI advancements have facilitated endotype classification and personalized therapeutic strategies, AI research in Asia may face challenges such as limited sample diversity and data heterogeneity, which hinder AI model generalizability. Therefore, methodologies for AI model training and validation have undergone continuous refinement, with enhanced techniques emerging in parameter optimization, cross-validation frameworks, and robustness verification protocols. While AI holds promising significance aiding in CRSwNP diagnosis and therapy, implementing AI in resource-limited settings may face clinical barriers, including infrastructure gaps (limited access to high-resolution imaging systems, hardware, etc.), lack of well-trained professionals, and integration challenges. Therefore, future studies should focus on multi-center collaboration, technical assistance based on cloud platforms and 5G network, and lightweight model development. Due to the differences in CRSwNP endotypes between the East and the West, the traditional "one-size-fits-all" approach is no longer valid. Differences in endotypes contribute to variations in responses to pharmacological treatment, surgery, and biologics. The flowchart guiding Asian clinicians in selecting treatments is presented in Figure 2. Although the efficacy and safety of T2 biologics have been proven in Western CRSwNP patients, due to heterogeneity in endotypes, the effectiveness of biologics in CRS may vary across different regions and genetic backgrounds. Therefore, multi-center clinical trials with large-scale Eastern patients are required to validate the applicability of T2 biologics in Asia. In addition, as for non-T2 CRSwNP characterised by neutrophilic or mixed endotype, prioritising real-world and clinical trial investigations into non-T2 biologics and small-molecule inhibitors is warranted, particularly across Asian populations. We believe that multicenter, international, and AI-assisted collaborations hold the potential to address a critical challenge in the management of severe CRSwNP. Junhai Chen and Linlu Wang were responsible for writing the main text of the manuscript; Xinyue Wang contributed to the manuscript structure and literature retrieval; Yana Zhang was responsible for revising and polishing the manuscript; Qintai Yang proposed the research topic and performed the final proofreading. The authors declare no conflicts of interest. Data sharing is not applicable to this article as no new data were created or analyzed in this study. Tables S1-S2. cea70150-sup-0001-TablesS1-S2.docx. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.