An intricate network of factors that interact at all disease levels drives asthma pathogenesis, from genetic susceptibility and immune system dysregulation to environmental factors' impact, epithelial barrier, mucus production, airway smooth muscle cells (ASM), extracellular matrix, local nerves and blood vessels. The central elements in asthma pathophysiology are chronic airway inflammation and tissue remodeling which can manifest through distinct pathways that define asthma endotypes. Type 2 (T2) immune response is the most studied and is initiated when the airway epithelial cells are fired upon by environmental factors (allergens, pollutants, and viruses) and release signal molecules known as alarmins [thymic stromal lymphopoietin (TSLP), IL (interleukin)-33 and IL-25], which activate innate lymphoid cells (ILC) type 2 (ILC2) and prone dendritic cells (DCs) to a T2 profile. Acting together, ILC2s and DCs recruit and activate T helper (Th) 2 cells with further secretion of IL-4, IL-13 IL-5, IL- 9 and IL-21 effector cytokines (15). IL-5 is involved in the recruitment, maturation and survival of eosinophils, while IL-13 is involved in goblet cell hyperplasia, mucus production and AHR. Together with IL-13, IL-4 induces the switch of B cells into IgE-secreting isotype plasma cells. Mast cells (MCs) and basophils are also recruited and activated (9, 11, 16–18). Beyond the T2 immune-inflammatory pathway, type 1 (T1) or type 3 (Th17) pathways may be activated in asthma. Environmental stimuli (pollutants, tobacco smoke, microbes), induce damage to airway epithelial cells activating the innate lymphoid cells type 3 (ILC3) and Th17 cells which secret pro-neutrophilic cytokines (IL-17 and IL-22). The Th1 pathway can be also involved with activation of ILC1, recruitment of CD4+ Th cells, CD8 + cytotoxic T cells, natural killer (NK) cells, and macrophages, inflammasome activation and production of interferon (IFN)-γ (9, 11, 19, 20).

Based on the predominant immune-inflammatory pathways involved major asthma endotypes include type-2 high asthma (T2 asthma), type-2 low (T1 and T3 asthma), and mixed endotypes (21, 22). These endotypes share a variety of pathophysiological pathways, including genetic and epigenetic, microbiome metabolic, neurogenic, and remodeling pathways (type V, VI and VII asthma) (8, 23). Non-type 2 pathogenetic pathways become noticeable especially in patients with severe T2 asthma, but might be found also in milder cases. A study involving adult patients with stable asthma showed overlapping T2 and non-T2 inflammation in 47.5% of the individuals evaluated (24, 25). A population Th2/Th17 dual-positive asthmatic patients was reported following the examination of bronchoalveolar lavage (BAL) fluid and was associated with more severe airway obstruction and AHR (13, 26, 27). In these patients, partial or no response to T2-targeted biologics can be explained by dominant non-T2 pathways driving the chronic inflammation, AHR and remodeling (24, 25). The contribution of non-T2 pathways within the T2 asthma endotype is supported by the efficacy of tezepelumab, an anti-TSLP monoclonal antibody, proved to reduce the exacerbations of asthma independent of blood eosinophil counts (28).

Biomarkers play a crucial role in linking asthma endotypes and phenotypes. Among the available biomarkers for T2 immune response in asthma, blood eosinophils, sputum eosinophils, and fractional exhaled nitric oxide (FeNO) are most frequently applied at the point-of care. Each has distinct advantages and limitations that influence their utility for endotype-driven decisions: (1) blood eosinophils are widely available, inexpensive, reproducible, predictive of exacerbation risk and response to anti–IL-5/IL-5R biologics, and they can be monitored longitudinally; however, blood eosinophil counts may be influenced by concomitant corticosteroid treatment, intercurrent infections, and various comorbidities, and the optimal cut-off values for clinical decision-making remain variable across studies (29, 30); (2) sputum eosinophils are considered the gold standard for assessing airway eosinophilic inflammation and correlate closely with disease control and future exacerbations; this parameter is also valuable in guiding inhaled corticosteroids (ICS) titration; however, the method requires specialized laboratory facilities and trained personnel, is technically demanding, and sample quality can be inconsistent, which limits its availability in routine clinical practice and its feasibility for repeated assessments (31–33); (3) FeNO provides a non-invasive, rapid, and patient-friendly assessment of T2 airway inflammation, reflecting IL-4/IL-13 activity, it is particularly useful in monitoring adherence to ICS and in predicting responses to therapies targeting IL-4Rα and IL-13; despite these advantages, FeNO measurements may be affected by age, atopy, smoking status, and respiratory infections, and elevated values are not specific to asthma, being observed in other eosinophilic airway diseases (34, 35). Taken together, these biomarkers provide complementary information. While sputum eosinophils offer the most direct measure of airway inflammation, blood eosinophils and FeNO are more accessible and practical for routine care. An integrated approach that combines these parameters with clinical context remains most informative for tailoring biologic therapy. They should be used within the regiotype and endotype variability framework. For instance, serum periostin, a biomarker commonly used to identify T2 asthma, has differing diagnostic value depending on ethnicity, age or asthma severity. While periostin effectively discriminates asthmatics from non-asthmatics in Japanese and Caucasian populations, it does not perform similarly in Chinese cohorts. Interestingly, Chinese individuals exhibit higher baseline periostin levels than Caucasians, and these levels also vary by sex within the Chinese population (5, 36). Although shown to be related to the T2 pathway in severe asthma periostin proved to be outside of the T2 cluster in patients with moderate asthma (37).

A recent hypothesis proposes that various aspects of asthma treatment response, such as lung function improvement, bronchodilator response, AHR, symptom control, oral corticosteroid (OCS) use, and emergency visits, are influenced by a single, quantitative corticosteroid responsiveness theratype (5). Genetic studies have identified several asthma pharmacotypes based on single nucleotide polymorphisms (SNPs) associated with responses to common asthma treatments (38, 39). Additionally, certain SNPs predict adverse effects of ICS in high doses, such as adrenal suppression (40, 41).

Currently, the endotype-driven strategies for asthma management use several biologic therapies targeting a specific pathogenetic pathway: anti IgE (omalizumab), anti IL-4/IL-13 (dupilumab), anti IL-5 (benralizumab, depemokimab, mepolizumab, reslizumab), TSLP (tezepelumab), IL-33/ST2 (itepekimab, tozorakimab, astegolimab) (14, 42, 43). Currently, 7 biologic agents are approved for asthma by the European Medicines Agency (EMA) and Food and Drug Administration (FDA) (Supplementary Table S1), with the majority targeting T2 asthma (44–46). Only tezepelumab can be used regardless of blood eosinophils.

The major clinical studies on EMA and FDA-approved biologics in asthma are summarised in Supplementary Table S2. Figure 2 presents a stepwise, endotype- and phenotype-driven algorithm to guide the selection of biological treatments in severe asthma, integrating key biomarkers, clinical characteristics, and practical considerations to support personalized therapy decisions.

Defining the response to targeted treatment in asthma remains a challenge and involves evaluating various clinical outcomes since there is no universally accepted definition. Recent studies showed that measuring the treatment response is complex and include reduced exacerbation frequency (at least a 50% reduction in the annual number of exacerbations), improved lung function (increase in forced expiratory volume (FEV₁) > 100 mL, OCS and ICS reduction, better symptom control and improved quality of life (QoL). Due to the variability of patient responses, the assessment should be performed using a comprehensive definition of response (47). Real-life data 50% of patients do not meet these criteria and are thus considered non-responders. Furthermore, even among responders, improvements may vary not covering all the clinical end-points (partial responders, dissociated response). The concept of “deep response”, encompassing not only exacerbation reduction but also symptom stability, exercise tolerance, and steroid independence, is gaining traction, particularly in real-world settings. Here, integrating objective measures (exacerbation rate, FEV₁, T2 biomarkers) with patient-reported outcomes (symptom burden, activity limitation, QoL) offers a more comprehensive framework for assessing the therapeutic benefit (48).

As no universally validated criteria for defining optimal response currently exist, several guidelines and consensus documents for biologics in severe asthma recommend incorporating shared decision making as a core element of response assessment (46, 49). The EAACI guidelines advocate for establishing individual, predefined treatment targets in partnership with each patient, based on their priorities and treatment goals, whether these focus on reducing exacerbations, minimizing corticosteroid use, improving QoL, or restoring functional capacity. This patient-centered approach is particularly relevant in a heterogeneous disease such as asthma, where treatment success may vary considerably between individuals.

Finally, the long-term disease modification and prevention of progression (e.g., airway remodeling, lung function decline) deemed as clinical, biological or complete remission is increasingly recognized an attainable target for endotype-driven asthma care (50).

CRS is increasingly recognized as a heterogeneous chronic inflammatory disorder rather than a single disease entity, with distinct pathogenic mechanisms driven by several immune-inflammatory pathways and complex remodeling processes. The understanding of CRS pathogenesis has evolved from a purely obstructive-infectious paradigm to a multidimensional model incorporating epithelial barrier dysfunction, type 1, 2 and 3 innate and adaptive immune response, mucus abnormalities, tissue remodeling, and microbial dysbiosis (53, 56).

The current endotypic classification of CRS defines subgroups according to the underlying immune mechanisms, such as T1 (IFN-γ-driven), T2 (IL-4/13, IL-5 or IgE driven) or T3 (IL-17-driven) (53, 55). Though challenging to implement at the point of care, endotyping represents the future of personalized therapy for CRS as it offers deeper insights into pathogenesis and guides targeted interventions (52, 57).

The T2 CRS endotype is the most well-characterized CRS endotype and is usually encountered in Western populations. It is characterized by a robust ILC2- and Th2 driven immune response, in which epithelial-derived cytokines TSLP, IL-33 and IL-25 initiate the T2 downstream signaling cascades (58). Notably, IL-5 and granulocyte-macrophage colony-stimulating factor (GM-CSF) promote eosinophil survival and activation, while IL-13 impairs epithelial barrier integrity by downregulating tight junction proteins such as claudins and occludin (59). T2-driven inflammation is closely associated with specific tissue remodeling patterns. These include basement membrane thickening, subepithelial edema, collagen deposition, and the accumulation of fibrin due to reduced fibrinolytic activity (53, 60). Downregulation of tissue plasminogen activator (tPA) and increased expression of coagulation components contribute to NP formation. Of note, remodeling in T2 CRS is actively modulated by cytokine-induced epithelial-mesenchymal transition (EMT), myofibroblast activation, and angiogenesis (53, 59).

T1 CRS endotype is defined by IFN-γ secretion from Th1 and by involvement of cytotoxic T lymphocytes (CTLs), macrophages and NK cells (56). This endotype is frequently associated with inflammasome activation and epithelial barrier disruption and has been reported to have increased frequency of intracellular pathogens or viral infection. CRSsNP, particularly in Asian populations (58), may exhibit a predominant T1 endotype. Eosinophils are sparse in this endotype. Elevated IL-12, TNF-α, and CXCL10 can serve as biomarkers for this endotype. Histologic biomarkers are epithelial denudation, mononuclear cell infiltration, and increased fibrosis. MMP-1 and IFN-γ are inversely correlated with the NP burden, suggesting protective or regulatory roles (55, 61).

T3 CRS endotype is defined by IL-17A and IL-17F secretion from Th17 and ILC3, is usually associated with extracellular bacteria and fungal infections (53). The T3 immune response involves neutrophil recruitment through IL-8, granulocyte colony-stimulating factor (G-CSF), and serum amyloid A (SAA). Recent evidence highlights the involvement of neutrophil extracellular traps (NETs), myeloperoxidase (MPO), and matrix metalloproteinases (MMP) 8, 9 and 3 as key mediators of T3-induced mucosal damage. The remodeling pattern in T3 CRS is distinct from T2 and T1, with excessive fibronectin deposition, basal lamina fragmentation, and increased expression of tissue inhibitors of metalloproteinases (TIMPs) (54, 59).

Staphylococcus aureus colonization is particularly important in CRSwNP, where its enterotoxins function as superantigens to amplify Th2 responses, drive local specific IgE synthesis, and impair host immunity. Additionally, S. aureus-induced IL-33 and TSLP production sustains chronic inflammation and tissue remodeling (62, 63).

Cluster-based analyses have refined the description of CRS endotypes. A landmark multicenter study in China identified five distinct clusters based on a combination of inflammatory and remodeling markers (56). Clusters 1 and 2 were characterized as non-T2, with high levels of IL-19 and IL-27, suggesting predominant regulatory or homeostatic profiles. IL-27 suppresses both T2 and Th17 responses and was inversely correlated with disease severity and recurrence. Cluster 3 was characterized by a low-T2/high-neutrophilic endotype, with the highest levels of MPO, IL-8, MMPs, and fibronectin. This endotype was associated with severe tissue remodeling and NET formation. Cluster 4 corresponded to moderate T2 inflammation, while cluster 5 exhibited the full T2 signature, including high levels of IL-5, GM-CSF, ECP, total IgE, and EETs (56).

CRS endotypes are influenced by environmental exposures—CRS regiotypes. In Western populations, CRSwNP is predominantly eosinophilic and T2-high, while in East Asia, CRSwNP often exhibits mixed T1/T3 features with low eosinophilia (58). Environmental exposures, microbial patterns, and genetic background likely contribute to these regional variations. For example, cluster 3 CRS in Chinese patients demonstrated significantly higher neutrophilic markers than those reported in European cohorts, emphasizing the need for region-specific endotyping strategies (56, 62, 64).

Clinically, T2 CRS is characterized by bilateral NP and frequent association with comorbidities such as asthma or non-steroidal anti-inflammatory drug (NSAID)-exacerbated respiratory disease (N-ERD). These patients often experience severe symptoms, particularly anosmia and nasal blockage, have high rates of disease recurrence post-surgery or need frequent OCS courses (65, 66). Non-T2 CRS typically presents without NPs and may be associated with pronounced tissue remodeling, including subepithelial fibrosis, goblet cell hyperplasia, and glandular hypertrophy.

Tissue eosinophilia, increased expression of periostin, eosinophil cationic protein (ECP) and CCL26 in NP tissue in and nasal secretions, the presence of eosinophil extracellular traps (EETs) and increased IgE levels, particularly specific IgE to microbial and self-antigens, serve as a robust signature biomarkers for the T2 CRS endotype (56). Blood eosinophilia and increased total serum IgE are frequently associated with T2 CRS, although the role of total IgE is questionable as it was not related to severity (67). Contrary to FeNO in asthma, nasal NO has not been shown to be helpful to identify the T2 endotype because the main source of production of nasal NO is the sinuses that are closed off when CRS occurs (68). In some more specialized centers tissue and serum periostin levels, IL-5 or IL-33 serum levels can be used (69, 70).

Neutrophils are typically predominant in 50% of patients with CRS without nasal polyps, but also are found to play a role in patients with severe type 2 CRS with nasal polyp disease. Programmed cell death ligand 1, platelet-derived growth factor subunit β [PDGF-β], macrophage inflammatory protein-3b, and PDGF-α were significantly predictive of the surgical outcome in non-eosinophilic CRS (71, 72).

The major theratypes of CRS are broadly categorized into T2 and non-T2 (73). This endotypic stratification aligns with principles of precision medicine and is increasingly integrated into clinical decision-making frameworks, such as those outlined in the 2020 European Position Paper on Rhinosinusitis and Nasal Polyps (EPOS 2020) (54, 74–76).

The therapeutic landscape for T2 CRS has been transformed by the introduction of biologic agents targeting key cytokines and immune pathways. The current EMA and FDA-approved biologics for CRSwNP are summarized in Supplementary Table S3; the major clinical studies on EMA-approved biologics in CRSwNP can be found in Supplementary Table S4. Dupilumab, an anti-IL-4Rα monoclonal antibody, has emerged as a first-line biologic for severe CRSwNP, demonstrating robust efficacy in reducing NP size, improving nasal airflow, and decreasing OCS dependence (77). Other approved agents include anti-IL-5 (mepolizumab, reslizumab) and anti-IgE (omalizumab) biologics. For T2 CRS, the integration of biomarker-driven algorithms supports the rational use of biologics, particularly in patients with bilateral NPs, elevated blood eosinophils or OCS dependence. The choice of the tailored intervention depends on biomarker profiles such as tissue eosinophil counts, blood eosinophil levels (≥250/μL), or serum IgE concentrations (≥100 IU/mL) (2, 54, 78).

Management of non-T2 CRS emphasizes infection control, mitigation of environmental exposures, and surgical intervention when appropriate (72, 74, 75). Patients with non-T2 CRS have a suboptimal response to OCS and are generally not candidates for currently available biologic therapies. Instead, long-term low-dose macrolide antibiotics, particularly clarithromycin, have been utilized for their anti-inflammatory and immunomodulatory properties, although high-quality evidence remains limited. Endoscopic sinus surgery (ESS) remains a mainstay for refractory cases, often serving both diagnostic and therapeutic purposes. A pragmatic, three-arm, randomised, placebo-controlled phase 4 trial, evaluated adults with CRS receiving ESS vs. clarithromycin or placebo. ESS proved clinical effectiveness, providing significantly improved disease-specific quality of life at 6 months. Conversely, the trial findings did not support routine long-term use of low-dose clarithromycin (79). Emerging investigational approaches include targeted inhibition of IL-17 and TNF-α pathways, although these are not yet validated for clinical use (5, 72).

Figure 3 presents a stepwise, endotype- and phenotype-driven algorithm to guide the selection of biological treatments in CRSwNP, integrating key biomarkers, clinical characteristics, and practical considerations to support personalized therapy decisions.

Defining treatment response in endotype-driven management combines clinical, radiologic, biomarker and patient-reported outcomes.

For T2 CRS response is assessed via nasal endoscopy (Lund-Kennedy/polyp scores), symptom scales (SNOT-22, VAS) and CT imaging (Lund-Mackay). EPOS 2020 outlines five key response criteria: polyp reduction, improved olfaction, fewer steroids, better QoL and comorbidity control, improvement in ≥3 defines success. Biologics like dupilumab show high efficacy in T2-high patients, particularly those with eosinophilia or asthma. Biomarker trends (e.g., decreased IgE, periostin) further guide therapeutic decisions, though transient eosinophilia post-dupilumab does not indicate failure. Non-T2 CRS requires distinct metrics, such as reduction in the neutrophilic markers (IL-8, MMP-9) over 6–12 weeks (54, 80).

Surgical response varies by endotype: T2 CRSwNP often recurs post-ESS without biologics, while non-T2 CRS may reach long-term relief.

CRS control frameworks (e.g., EUFOREA) now emphasize stability, exacerbation rates, and PROMs over isolated metrics (80). Emerging tools, such as AI-guided endoscopy and nasal biomarker signatures, hold the promise for real-time monitoring of the efficacy of targeted interventions.

Given the heterogeneity of CRSwNP and its frequent overlap with comorbid conditions, shared decision making is likewise essential when establishing individualized treatment goals for biologic therapy. Targets such as reduction in NP size, symptom relief (nasal obstruction, loss of smell), decreased need for OCS or surgery, and improved QoL should be defined in partnership with the patient, ensuring that response assessment reflects outcomes meaningful to them (54, 80).

Ultimately, CRS management is evolving toward dynamic, personalized algorithms, integrating biomarkers, biologics and advanced diagnostics for precision care.