Introduction

Chronic rhinosinusitis with nasal polyposis (CRSwNP) represents a clinical phenotype of chronic rhinosinusitis (CRS) characterized by persistent polypoid inflammation of the sinonasal mucosa [1]. Despite the fact that CRSwNP accounts for only approximately 25–30% of all CRS cases, CRSwNP is considered a more severe condition that seriously affects patient’s quality of life (QOL) and productivity, and is associated with a large economic burden [2]. Unfortunately, significant variations in clinical manifestations, therapeutic responses, and prognosis in patients with CRSwNP make this disease difficult to identify, assess, and manage. This has strengthened the concept that rather than being one single disease entity, CRSwNP appears to be a heterogeneous disease spectrum consisting of several different endotypes that might be discerned by corresponding biomarkers [3]. For these reasons, there is a clear need for reliable biomarkers to provide additional information beyond that supplied by conventional clinical measures.

In recent years, basic research has shed some light on the pathogenesis of CRSwNP and has contributed to the discovery of various biomarkers from airway inflammation and tissue remodeling. Compared with clinical traits, endoscopic examinations, and imaging techniques, biomarkers are more objective, quantitative and reflective of the underlying pathophysiology [4]. Successful identification and clinical translation of CRSwNP biomarkers can aid to classify patients, monitor treatment effects, assess risk of recidivism, and develop novel biological therapies. In this paper, we review the current state of knowledge with respect to biomarkers in the evaluation and management of CRSwNP.

Current sources of CRSwNP biomarkers

In general, CRSwNP biomarkers can be assessed in the nasal secretions, nasal biopsies, exhaled breath via nasal nitric oxide (nNO) evaluation, and peripheral blood. Soluble mediators such as chemokines and cytokines can be obtained from nasal secretions. In addition, it has been shown that cytokines in nasal secretions typically correlate with tissue levels [5]. Nasal biopsies can provide more detailed and accurate information about structural components, cellular patterns and expression of inflammatory mediators. Similarly to the lower airways, many recent studies have evaluated sinonasal diseases by measuring nNO, which is a simple, rapid, well-tolerated, and noninvasive method [6]. In the clinical setting, the identification of biomarkers in peripheral blood is relatively cost-effective and readily available.

Biomarker discovery in CRSwNP is a steadily developing field, and different kinds of biomarkers have been explored in well-defined patient cohorts. At present, most of the literature on CRSwNP biomarkers is concerned with proof-of-concept studies of single markers which have been deduced from related biological processes in other diseases such as asthma and atopic dermatitis, or have been identified from genomic and proteomic data. For instance, Liu et al. [7] discovered 192 upregulated and 156 downregulated disease-related genes in polyp tissue by DNA microarray technology. Using standard proteomic methodology, Upton et al. [8] identified more than 300 differentially expressed proteins in sinonasal mucosa of CRSwNP patients. Theoretically, all biological components participating in pathophysiological processes could be eligible candidate biomarkers. Nevertheless, each candidate is involved in different molecular pathways and may capture only a fraction of the information. Consequently, a multiplex panel of biomarkers will be required to better clarify such complex pathological processes and to improve their predictive power.

Biomarkers for CRSwNP evaluation and management

In the context of pathogenic mechanisms, CRSwNP biomarkers appear to fall into several main categories, including eosinophils, type 2 cytokines, immunoglobulins, remodeling factors, and nNO, along with molecules involved in steroid responsiveness and olfactory loss (Fig. 1).

Fig. 1
figure 1

Schematic overview of various biomarkers involved in the pathophysiology of CRSwNP. On exposure to pathogens or allergens, epithelial cells (ECs) can release TSLP, IL-25, and IL-33, which subsequently act on ILC2s to produce IL-5 and IL-13. In addition, TSLP stimulates DCs to induce Th2-cell differentiation with an amplified type 2 cytokine response. Furthermore, IL-5 is responsible for promoting eosinophil recruitment, activation, and survival. IL-4 and IL-13 not only promote basilar secretion of periostin from ECs, but also contribute to local production of antibodies, including several autoantibodies in severe cases. In addition, NO is constantly being released in the nasal airways. Abnormal expression of MUC1 and MUC4 in ECs may cause steroid resistance. MUC1, Mucin 1; MUC4, Mucin 4; nNO, nasal nitric oxide; TSLP, thymic stromal lymphopoietin; DC, dendritic cell; ILC2, type 2 innate lymphoid cell

Full size image

Tissue and blood eosinophils

Classically, eosinophils are considered key effector cells infiltrating the polyp tissue. However, more than 50% of CRSwNP cases in East Asia present noneosinophilic inflammation [9, 10]. In general, eosinophilic nasal polyps (NPs) are pharmacologically glucocorticoid responsive, but noneosinophilic or neutrophilic CRSwNP is frequently resistant to steroid treatment [11–13]. Moreover, tissue eosinophilia has been recognized as a valuable predictor of augmented disease severity and worse prognosis after endoscopic sinus surgery (ESS) [14, 15]. Hence, several groups have argued endotyping patients into eosinophilic versus noneosinophilic subtypes to establish a personalized treatment plan. Currently, there is no unanimous histopathologic criterion to exactly define tissue eosinophilia. Kountakis et al. [16] proposed a cut-point of >5 eosinophils per high-power field (HPF) based on the degree of activated eosinophils and patients with tissue eosinophilia had more severe disease as measured by computed tomography (CT) and endoscopy scores. In a study by Soler et al. [17], mucosal eosinophilia was defined as >10 eosinophils/HPF, because this was predictive for less QOL improvement after surgery. More recently, Lou et al. [18] enrolled 387 Chinese patients with CRSwNP in a cohort and demonstrated that a tissue eosinophil proportion of >27% or an absolute tissue eosinophil count of >55 eosinophils/HPF predicted the relapse of NPs within 2 years after ESS. Tokunaga et al. [19] used a larger cohort of patients in Japan and found that tissue eosinophilia of more than 70 eosinophils/HPF was strongly associated with recurrence after surgery.

Blood eosinophilia is also correlated with CRS outcomes. Matsuwaki et al. [20] reported that CRS patients with peripheral eosinophil count above 520/μl were more likely to experience relapse within 5 years after ESS. As an alternative, blood eosinophil levels have been investigated as a potential surrogate marker for tissue eosinophilic inflammation in CRS and are easy to obtain in an outpatient setting [14]. In one study [21], a blood eosinophil percentage of ≥6% predicted eosinophilic CRS with a sensitivity of 97.4% and a specificity of 70.7%. In another study [22], a blood eosinophil number of 0.16 × 109/L yielded a sensitivity of 84.9% and a specificity of 84.4% and a blood eosinophil ratio of 2.05% yielded a sensitivity of 89.0% and a specificity of 84.4% for the diagnosis of eosinophilic CRS. Of note, various disorders and causes, including allergies, autoimmune diseases, adverse drug reactions and parasitic infections, can alter circulating eosinophil counts [23]. Accordingly, blood eosinophilia does not necessarily reflect tissue eosinophilia and its utility remains limited.

Type 2 cytokines as promising therapeutic targets

IL-5 is a crucial driver for eosinophil recruitment, activation, and survival [24]. In patients with CRSwNP, it has been confirmed as the main positive determinant of eosinophilic inflammation [25]. IL-5 positive NPs are associated with a higher likelihood of suffering from comorbid asthma and revision surgery [25–27]. Furthermore, a recent cluster analysis generated 10 inflammatory endotypes of CRS based on an array of biomarkers, and IL-5 was a critical parameter to dictate the phenotype with NPs and asthma [28]. Therefore, IL-5 plays an important role in CRSwNP pathogenesis and anti-IL-5 treatment may be a promising therapeutic strategy in these patients. Indeed, two clinical trials have examined the potential of IL-5 antagonism in cases with severe CRSwNP. Reslizumab and mepolizumab, which are both humanized antibodies directed against IL-5, significantly reduced the size of NPs and the number of blood eosinophils [29, 30]. Interestingly, IL-5 levels in nasal secretions were found to predict the response to treatment with reslizumab [29], thus emphasizing the importance of biomarkers for the appropriate patient selection.

IL-4 and IL-13 are also essential cytokines that activate type 2 inflammatory responses. They synergistically promote the synthesis of IgE from B cells and stimulate mucus secretion in epithelial cells (ECs). Despite the lack of association between these two cytokines and conventional measures of disease burden, blocking the IL-4/IL-13 signaling pathway has become a major step forward in the management of CRSwNP [31]. In a recent Phase II clinical trial, dupilumab, a fully human monoclonal antibody to the α subunit shared by IL-4 and IL-13 receptors, significantly improved clinical, endoscopic, radiological, and pharmacodynamic outcomes in patients with nasal polyposis after 16 weeks [32].

In addition, there is growing evidence that newly identified epithelial-derived cytokines, including IL-25, IL-33, and thymic stromal lymphopoietin (TSLP), have the capacity to shape the type 2 inflammatory milieu via effects on dendritic cells (DCs), mast cells, and group 2 innate lymphoid cells (ILC2s), as well as adaptive Th2 cells [33–36]. IL-25, also known as IL-17E, is a member of the IL-17 cytokine family. It has been reported that increased IL-25 expression is associated with poor CT scores and elevated blood eosinophil numbers in CRS patients [37]. In a murine NP model, IL-25 blockade not only reduced the number of polypoid lesions and the thickness of edematous mucosa, but also suppressed infiltration of inflammatory cells and expression of associated chemokines and cytokines, thus indicating the possibility of IL-25 as a novel pharmacologic target for CRSwNP patients [38].

IL-33, which is a nuclear cytokine from the IL-1 family and a ligand for the orphan IL-1 family receptor ST2, plays a pathogenic role in several Th2-related inflammatory diseases [39, 40]. Inhibition of the IL-33/ST2 signaling pathway has been shown to ameliorate eosinophilic airway inflammation and to reduce Th2 cytokine production in a murine model of allergic asthma [41]. In the case of CRS, one study reported that baseline expression levels of IL-33 were markedly increased in treatment-recalcitrant patients with CRSwNP [42]. Another study found that IL-33 concentrations were negatively associated with CT findings in CRS patients [43]. Intriguingly, in a recent study by Kim et al. [44], IL-33 was related to the expression of various Th1/Th17 cytokines and neutrophil recruitment in Asian patients with CRSwNP. Moreover, in a murine model of CRS, IL-33 neutralizing antibody diminished mucosal thickness and subepithelial collagen deposition, and inhibited infiltration of neutrophils. These findings indicate that IL-33 may represent a potential new therapeutic target for noneosinophilic CRSwNP.

TSLP, an IL-7-like cytokine, appears to facilitate Th2-skewed responses and subsequent recruitment of eosinophils in the airways. Enhanced expression and activity of TSLP have been found in polyp tissue [45]. Furthermore, TSLP expression positively correlated with the markers of Th2-biased eosinophilic inflammation in sinonasal mucosa and with symptom and CT scores in eosinophilic CRSwNP [46]. More importantly, in a Phase II double-blind, placebo-controlled trial, AMG 157, a human monoclonal antibody binding to TSLP, has been shown to attenuate allergen-induced early and late asthmatic responses and to reduce sputum and blood eosinophils in patients with allergic asthma [47]. Since TSLP also contributes to type 2 immunity in CRSwNP, neutralizing TSLP may become an effective pharmacologic method in patients with eosinophilic CRSwNP.

B-cell-associated biomarkers

B cells are one of the most important components of adaptive immunity. It is becoming increasingly apparent that B cells may play an essential role in perpetuating inflammatory reactions observed in patients with CRSwNP [48]. The numbers of naïve B cells and activated plasma cells are significantly upregulated within nasal polyp tissue [49]. In addition to their ability to secrete pro-inflammatory mediators and function as antigen-presenting cells for T cells, B cells can produce a variety of antibodies at mucosal sites. It has been well documented that levels of many different isotypes of immunoglobulins are highly elevated in NPs, but not in sera, of patients with CRSwNP, indicating a local immune response rather than a systemic response [49].

Mucosal tissue polyclonal IgE, as a major contributing factor in CRSwNP, has already been studied with some showing promising results [50]. Bachert et al. [51] found that concentrations of total IgE correlated significantly with levels of eosinophil cationic protein (ECP) and IL-5, and thus local eosinophilic inflammation. In a prospective follow-up study performed over 12 years after ESS, Gevaert et al. [26] showed that high local IgE could predict whether a patient was likely to experience a recurrence requiring repeated surgery. Furthermore, in patients with CRSwNP, tissue IgE and specific IgE against S. aureus enterotoxins (SE-IgE) were identified as risk factors for the comorbidity of asthma [25]. Finally, in a recent proof-of-concept clinical trial, omalizumab, a recombinant humanized monoclonal antibody to IgE, demonstrated significantly improvements in clinical parameters, symptoms, and QOL measures in patients with CRSwNP and comorbid asthma, irrespective of their allergic status [52]. Therefore, IgE may be an attractive therapeutic target in a specific subset of patients with CRSwNP.

In addition, the levels of several autoantibodies have been reported to be increased in CRS patients. Tan et al. [53] showed that IgG and IgA autoantibodies against nuclear antigens were highly elevated in polyp tissue, and anti-dsDNA IgG antibody was strongly correlated with a more severe clinical course necessitating a second surgical intervention. In addition, Goncalves et al. [54] detected antineutrophil cytoplasmic antibodies (ANCA) in serum samples in 49 consecutive CRS patients, and concluded that ANCA might be a helpful clue in establishing the early diagnosis of undisclosed vasculitides in a subset of difficult-to-treat CRS patients. Collectively, the presence of several specific autoantibodies insinuates a more aggressive form of CRS and may be used as clinically useful biomarkers.

Markers of tissue remodeling

Besides the persistent airway inflammation, tissue remodeling is another cardinal pathologic feature of CRSwNP. It is a dynamic process of formation and degradation of extracellular matrix (ECM), leading to transient or permanent changes in tissue architecture [55]. In CRSwNP, it is exhibited in both the mucosa and bone, characterized by epithelial damage and erosions, basement membrane thickening, massive stromal edema, pseudocyst formation, and osteitis [56]. Matrix metalloproteinases (MMPs) are key enzymes with proteolytic activities that can degrade various components of ECM and contribute to progressive histologic changes. Among the MMP family, MMP-9 levels have been extensively reported to be increased in NPs [57–60]. However, no significant correlation has been demonstrated between MMP-9 expression and the severity of CRSwNP assessed by CT scans and polyp grades [58, 60]. Furthermore, Watelet et al. [61, 62] have found that MMP-9 concentrations in nasal fluids are linked to MMP-9 expression inside the ECM, and high MMP-9 amounts in the pre- and postoperative period predict unfavorable healing outcome after ESS. In addition, MMP-9 may also serve as an interesting therapeutic target in CRSwNP. Doxycycline possesses an anti-MMP effect, and has been shown to significantly lower MMP-9 levels in secretions, modify polyp size, and ameliorate postoperative healing quality [63, 64].

Periostin is a secreted extracellular protein that has emerged as a marker in pathologic remodeling process. Periostin is produced mainly by ECs in response to IL-4 and IL-13 and has a pivotal role in subepithelial fibrosis [65, 66]. Periostin expression has been confirmed to be increased in patients with CRSwNP, independent of their atopic status or comorbid asthma [65]. Elevated periostin in the sera and nasal fluids may be used to differentiate between patients with CRSwNP and healthy subjects or patients with allergic rhinitis [67]. Moreover, Zhang et al. [68] found that periostin expression was elevated in active CRS, and decreased after effective treatment, suggesting that periostin could represent an indicator of disease activity and responsiveness to therapy. On the other hand, periostin has been perceived as a marker of Th2-type eosinophilic inflammation. Tissue periostin levels were significantly associated with IL-5 expression [65]. In eosinophilic CRSwNP, periosin levels positively correlated with radiographic findings [69].

Nasal nitric oxide

Nitric oxide is a regulatory mediator widespread in the body. In the nasal airways, it is produced constantly from ECs lining the paranasal sinuses, and contributes to local host defense by bacteriostatic and antiviral properties and by stimulation of ciliary motility [6]. Numerous reports have so far supported a role for nNO in CRSwNP assessment and management. A frank decrease in nNO has been observed in patients with nasal polyposis, mainly because of sinus ostial occlusion and failure of gaseous NO in the sinuses to reach the nasal cavities [70, 71]. Testing for nNO might be valuable for the distinction of subjects having NPs from those without NPs. Jeong et al. [72] reported that the cut-point of less than 163 ppb was suggestive of CRSwNP with a sensitivity of 81.3% and a specificity of 93.3%. In symptomatic patients suspected of having CRS, nNO cut-off value of <442 ppb could allow for screening and diagnosis of patients suffering from CRSwNP with a sensitivity of 87% and a specificity of 91% [73].

In addition, in cases of CRSwNP, the initial nNO levels correlated inversely with the extent of sinus disease as documented by CT changes and endoscopic appearances, and might fill the niche for an objective and noninvasive measure of NP severity [70–74]. Following surgical or medical treatment, the levels of nNO were dramatically elevated in parallel with improvements in an array of clinical markers of disease activity [71, 74, 75]. Hence, the change in nNO over time could help to assess the patency of sinus ostium and to determine the effectiveness of therapeutic interventions.

Biomarkers of corticosteroid responsiveness

Glucocorticoids (GCs) remain the mainstay of CRS treatment and are most effective drugs to control NP inflammation and growth. However, the response to GCs is variable and a proportion of patients exhibit steroid insensitivity [76]. Given these observations, biomarkers used for identification of GC-resistance may allow more tailored therapy, while minimizing inappropriate use and undesirable side-effects. As previously discussed, Wen et al. [11] revealed that increased accumulation of neutrophils in NPs was correlated with corticosteroid hyporesponsiveness. In addition, glucocorticoid receptor β (GRβ), an endogenous inhibitor of GC action, appears to be linked to steroid treatment response in CRSwNP [77, 78]. In one study, Hamilos et al. [79] reported an inverse relationship between baseline GRβ expression in NP inflammatory cells and the efficacy of topical fluticasone. In another study, Valera et al. [80] found higher expression of GRβ in poor responders following 2-month treatment with intranasal budesonide.

Recently, studies on the anti-inflammatory effects of GCs demonstrated that the abnormal expression level of mucin 1 (MUC1) and mucin 4 (MUC4) might be functionally involved in steroid resistance. Milara et al. [81] showed that MUC1 expression was significantly downregulated in NP epithelium of CRSwNP patients resistant to systemic corticosteroids. Opposite results were observed for MUC4 expression [82]. Thus, the higher ratio of MUC4/MUC1 in patients with CRSwNP could help to distinguish poor versus favorable responders.

Biomarkers for assessment of olfactory function

Olfaction is an important sensory function and is deeply involved in appetite, motivation, and libido. Olfactory loss represents a frequent complaint for patients with CRS, especially with CRSwNP, and has a significant long-term impact on QOL [2, 83]. It has been reported that eosinophilic CRS can induce more severe smell impairment even in early stages [12]. In a study by Mori et al. [84], serum total IgE ≥ 400 IU/ml was identified as an independent risk factor for olfactory loss in patients with eosinophilic CRS. In addition, treatment with anti-IgE agents could contribute to a significant improvement in subjective olfactory function [52]. More recently, it has been found that increased levels of eosinophils within the olfactory tissue were strongly related to olfactory disturbance in CRSwNP patients, implying that local eosinophils might act as essential effectors in olfactory dysfunction [85]. Schlosser et al. [86] examined correlations between an array of inflammatory mediators in mucus from the olfactory cleft and objective olfactory function assessed using the Sniffin’ Sticks test, and found that IL-5 levels correlated inversely with olfactory scores irrespective of polyp status. In addition, neuron-specific enolase (NSE), as an indicator of neuronal injury, has been reported to be elevated in nasal secretions in CRSwNP patients, and the higher concentrations of NSE were linked to worse olfactory function [87].

Conclusion

The heterogeneous nature of CRSwNP is progressively being recognized and acknowledged, necessitating the discovery, refinement, and utilization of biomarkers to delineate specific endotypes, monitor disease activity, and determine the optimized therapeutic modality. Currently, various potential biomarkers have been explored in nasal secretions, nasal biopsies, exhaled air, and peripheral blood. Tissue eosinophil count represents a valuable marker for endotyping, assessing disease severity, predicting steroid responsiveness and NP recurrence after ESS. Blood eosinophilia is related to tissue levels, whereas its utility remains limited. Type 2 cytokines and IgE have been proposed as essential targets for biological therapies. Furthermore, MMP-9 is correlated with postoperative healing quality. Measuring nNO may serve as a noninvasive tool for assessment of sinus ostial patency. In addition, MUC1 and MUC4 are novel biomarkers involved in steroid resistance. NSE appears to be a useful indicator for olfactory dysfunction. Despite these promising results, very few of biomarkers have emerged with sufficient power and enough evidence to support their feasibility and reproducibility in routine clinical practice. Longitudinal and prospective studies are needed to better characterize the role of biomarkers, individually or in combination, in the evaluation and management of CRSwNP.