Osteopontin as a Novel Biomarker in Distinguishing Chronic Rhinosinusitis with Nasal Polyp Endotypes and Predicting Disease Severity Free

Chronic rhinosinusitis (CRS) represents a prevalent chronic inflammatory condition affecting the nasal mucosa, characterized by a disease duration exceeding 12 weeks [1‒3]. Recent epidemiological reports have indicated that the prevalence of CRS ranges from 5.5% to 28% among the global population, with an ongoing trend of increasing incidence [4]. The condition can be categorized into two distinct phenotypes: CRS with nasal polyps (CRSwNPs) and CRS without nasal polyps. Patients diagnosed with CRSwNP exhibit more pronounced clinical symptoms, a higher rate of relapse, and an elevated likelihood of comorbidities, such as asthma and allergic rhinitis [5, 6]. Furthermore, CRSwNP is categorized into two distinct endotypes: eosinophilic CRSwNP (eCRSwNP) and non-eosinophilic CRSwNP (neCRSwNP). The diagnostic criteria predominantly rely on the quantitative assessment of eosinophilic infiltration through histopathological examination [7‒9]. Each endotype exhibits unique clinical characteristics, therapeutic approaches, and prognostic outcomes. Consequently, there is a pressing demand for the discovery of objective markers or biomarkers capable of distinguishing between the endotypes of CRSwNP, thereby facilitating the development of targeted treatment protocols and the monitoring of therapeutic efficacy.

Osteopontin (OPN), a phosphorylated glycoprotein integral to the extracellular matrix, exhibits a molecular weight of approximately 44 kilodaltons and is composed predominantly of amino acid residues, the secondary structure of polypeptide chains, and oligosaccharide moieties [10]. Ubiquitously distributed across various tissues, cells, and bodily fluids, OPN displays affinity for a range of integrins [11]. OPN exerts a pivotal influence on airway remodeling through its regulatory effects on TGF-β1, various endothelial growth factors (VEGFs), and Th2 cytokines (such as IL-4, IL-5, and IL-13) [12, 13]. Recent studies have found that OPN and VEGF are overproduced in CRSwNP, and OPN induces VEGF production, suggesting that the OPN-VEGF axis may be involved in CRSwNP angiogenesis [14]. Eosinophil-derived OPN has been shown to activate nasal fibroblasts, thereby facilitating inflammation and tissue remodeling in eCRSwNP [15]. Considering the role of OPN in eosinophil-mediated inflammatory processes, OPN in nasal secretions may herald a promising noninvasive and highly patient-compliant diagnostic strategy. This approach demonstrates significant potential in preoperatively distinguishing between eCRSwNP and neCRSwNP, which is crucial for devising personalized treatment strategies and subsequent monitoring regimens tailored to the specific endotype.

A cohort comprising 32 healthy controls (HCs) and 81 individuals diagnosed with CRSwNP, who were subjected to endoscopic nasal surgery at our medical institution during the period of January 2023 to June 2023, was assembled. The diagnosis of CRSwNP was established in accordance with the guidelines stipulated by the EPOS 2020 [16]. Subjects presenting with fungal sinusitis, sinus tumors, autoimmune disorders, or eosinophilic conditions were excluded from the study. Patients were not administered immunomodulatory agents, antibiotics, corticosteroids (neither topical nor systemic), or anti-allergic medications within the 4-week period preceding the surgical intervention. Demographic and clinical parameters of all enrolled subjects were meticulously documented. Subjective symptomatology was comprehensively evaluated utilizing the Visual Analog Scale (VAS). The preoperative computed tomography was scored in accordance with the Lund-Mackay scoring system, while the preoperative nasal endoscopy was appraised using the Lund-Kennedy scoring system. A control group, comprising thirty-two HCs exhibiting no clinical signs of rhinitis, sinusitis, or any inflammatory or autoimmune conditions, was established for comparative analysis within this study. This study was approved by the Medical Ethics Committee of Renmin Hospital of Wuhan University (No. WDRY2022-K227), and all enrolled patients signed informed consent.

The expansive sponge was inserted into the middle nasal meatus of the patient and placed for 30 min before removal. The operation was gentle without damaging the nasal mucosa or bleeding. The nasal secretions in the expansive sponge were squeezed into the centrifuge tube. After centrifugation at 4°C and 3,000 revolutions per minute for 15 min, the supernatant was placed in a cryopreservation tube and stored at –80°C. Utilizing an ELISA kit procured from Shanghai Enzyme-linked Biotechnology Co., Ltd., the levels of OPN in the nasal secretions were quantified in strict accordance with the manufacturer’s protocols. The optical density of each sample was ascertained at a wavelength of 450 nanometers using a spectrophotometer.

Nasal polyp specimens were procured and subjected to fixation in a 10% formalin solution, followed by paraffin embedding. Subsequently, the embedded tissues were sectioned and stained using hematoxylin and eosin to facilitate the microscopic examination of inflammatory cellular infiltrates. Utilizing a microscope with a ×400 magnification, the assessment of eosinophils was conducted within five randomly selected high-magnification fields by two independent, blinded evaluators devoid of clinical bias. The average count of eosinophils was meticulously documented. A tissue eosinophil count exceeding 10% was employed as the criterion for the classification of eCRSwNP, whereas specimens with a percentage below this threshold were designated as neCRSwNP [7].

In compliance with the guidelines established by the American Thoracic Society/European Respiratory Society guidelines, nasal fractional exhaled nitric oxide (FnNO) levels were assessed prior to therapeutic intervention utilizing a nitric oxide analyzer manufactured by Sunvou Corporation, Wuxi, China. Prior to the assessment, participants were instructed to abstain from smoking, engaging in strenuous physical activity, consuming nitrogen-rich foods, such as organ meats, lettuce, and spinach, and the intake of caffeinated or alcoholic beverages for a duration of 1 h. Following a period of repose lasting 30 min, subjects were positioned in a seated posture with an olive-shaped nasal occluder inserted into a single nostril. Subjects were then instructed to maintain a slight closure of the mouth and to engage in normal respiratory activity. The apparatus continuously extracted exhaled gases through a sampling conduit, maintaining a flow rate of 5 mL per second, thereby facilitating the automated analysis of the FnNO concentration. To ensure accuracy, the measurement protocol mandates the execution of at least three trials, with the resultant FnNO values being averaged.

After dewaxing, the sections were repaired by antigen and incubated with goat serum for 30 min. The anti-OPN antibody (1:400, Proteintech) was incubated overnight at 4°C. On the second day, biotin-labeled goat anti-rabbit IgG was added and incubated at room temperature. After washing with PBS, DAB was used for color development and neutral gum was used to seal the slices. Then, the staining intensity was evaluated by ImageJ software.

All results were represented as mean ± SEM. The data and graphs were analyzed by GraphPad Prism 8.0. The multiple groups were analyzed by one-way ANOVA followed by post hoc Tukey’s tests for multiple comparisons, while an unpaired Student’s t test was utilized for binary comparisons. Categorical data were expressed as numbers (%) and were compared using the χ² test. To discern potential predictive biomarkers for eCRSwNP, Pearson correlation analysis was implemented. The predictive efficacy of the identified genes was assessed using the receiver operating characteristic (ROC) curve analysis. The Youden index, formally defined as the sum of sensitivity and specificity subtracted by one, serves as a critical metric for ascertaining the optimal cutoff point in diagnostic testing. This index is maximized at the threshold where the combined discriminatory power of sensitivity and specificity is at its zenith, thereby delineating the most effective cutoff for binary classification. A p value <0.05 was considered significant.

Based on histological assessment, a cohort of 81 patients diagnosed with CRSwNP was stratified into two distinct groups: the neCRSwNP and the eCRSwNP. Demographic and clinical characteristics for the neCRSwNP, eCRSwNP, and the HC groups are delineated in Table 1. Representative HE-stained images of neCRSwNP and eCRSwNP are shown in Figure 1. Notably, the eCRSwNP group exhibited a significantly elevated prevalence of allergic rhinitis and asthma, as well as increased blood eosinophil counts, blood eosinophil percentage, and serum total IgE levels in comparison with the other groups. It is noteworthy that no substantial differences were observed in terms of age, gender, smoking status, VAS scores, Lund-Mackay scores, and Lund-Kennedy scores across the three groups (p > 0.05).

To confirm the role of OPN in CRSwNP, an investigation was conducted to assess the levels of OPN in both tissue samples and nasal secretions. Compared with the control group, the OPN levels in the tissues of eCRSwNP and neCRSwNP groups were significantly increased, and the OPN levels in eCRSwNP group were higher than those in neCRSwNP group (Fig. 2a, b). Similarly, the level of OPN in nasal secretions in CRSwNP group was found to be significantly elevated relative to the HC group, with the eCRSwNP group demonstrating an even more pronounced increase over the neCRSwNP group (Fig. 2c, d). The Pearson correlation analysis indicated a robust correlation between the OPN levels in nasal secretions and those in the corresponding tissue samples (Fig. 2e).

To assess the association of OPN levels in nasal secretions with factors associated with CRSwNP endotypes, Pearson correlation analysis was conducted. A positive association was identified between elevated OPN levels in nasal secretions and several eosinophil-related metrics, including blood eosinophile count (r = 0.715, p < 0.001) and percentage (r = 0.353, p = 0.001), tissue eosinophile count (r = 0.479, p < 0.001), and VAS score (r = 0.266, p = 0.016), Lund-Mackay score (r = 0.298, p = 0.006), or Lund-Kennedy score (r = 0.239, p = 0.031). However, no significant correlations were observed with serum total IgE (r = 0.182, p = 0.103) and FnNO (r = 0.194, p = 0.082) (Fig. 3). Furthermore, the ROC curve analysis revealed that the predictive accuracy of OPN levels in nasal secretions (AUC = 0.850, p = 0.001) for CRSwNP endotypes was superior to that of blood eosinophile count (AUC = 0.759, p < 0.001), blood eosinophile percentage (AUC = 0.510, p = 0.872), serum total IgE (AUC = 0.652, p = 0.018), FnNO (AUC = 0.594, p = 0.145) (Fig. 4).

Utilizing the maximum Youden index of 0.601 as a reference point, the calculated sensitivity and specificity for the prediction of eCRSwNP were determined to be 81.82% and 78.38%, respectively. Correspondingly, the established cutoff value for OPN levels in nasal secretions was identified as 121.05 ng/mL. Subsequent immunohistochemical analysis corroborated that elevated OPN concentrations in nasal secretions were indicative of increased OPN expression within the tissue samples (Fig. 5). Furthermore, the ROC curve analysis demonstrated that OPN concentrations in nasal secretions provided a more precise prediction of the VAS score, Lund-Mackay score, and Lund-Kennedy score, as compared to the endotype classification based on tissue eosinophil counts (Fig. 6).

Numerous scholarly investigations have reported a heightened prevalence of asthma among individuals with eCRSwNP, as well as a pronounced tendency for postoperative recurrence [7‒9]. The current diagnostic criteria for eCRSwNP predominantly hinge on the quantitative assessment of eosinophilic infiltration through histopathological examination. According to the EPOS 2020 guidelines, eCRSwNP is characterized by an eosinophil count exceeding 10/high power field at ×400 magnification or higher [16]. Cao et al. [7] have delineated eCRSwNP as instances where the eosinophil proportion in nasal mucosa or tissue exceeds 10%, in contrast to neCRSwNP, where this proportion falls below 10%. This classification system facilitates the exploration of eosinophilic inflammatory mechanisms within CRS, yet it does not bear a direct association with the disease’s prognosis. Notably, discernible disparities have been identified between the two endotypes in terms of clinical manifestations, symptomatology, prognostic indicators, and the incidence of postoperative relapse [5]. Consequently, the preoperative differentiation of eCRSwNP from neCRSwNP, coupled with the postoperative anticipation of recurrence, is imperative for the formulation of tailored therapeutic strategies and the enhancement of subsequent monitoring regimens. The analysis of biomarkers in nasal secretions emerges as a noninvasive diagnostic approach with considerable potential, attributable to its ease of execution, high patient compliance, and accurate reflection of localized inflammatory processes [17, 18]. In our study, a robust correlation was observed between OPN levels in nasal secretions and those within tissue samples, thereby implying that OPN levels in nasal secretions could potentially be employed as a noninvasive prognostic indicator.

Previous studies have substantiated the pivotal role of OPN in a spectrum of physiological processes, including tissue remodeling, fibrosis, inflammation, immune response, and the progression of malignancies, notably tumor invasion and metastasis [11]. In the context of airway remodeling, OPN exerts its influence by modulating the interaction between TGF-β1 and a variety of endothelial growth factors, such as VEGF [13]. Correlative studies have consistently demonstrated a positive association between the thickness of the airway basement membrane and the expression levels of OPN in asthmatic patients [19]. Moreover, the serum concentration of OPN n pediatric asthma patients has been found to correlate positively with TGF-β1 levels, which are implicated in the thickening of the airway wall and the activation of airway epithelial cells, thereby facilitating airway remodeling through the promotion of TGF-β1 expression [20]. In experimental models utilizing ovalbumin-induced asthma in mice, an upregulation of OPN expression was observed in lung tissue and bronchoalveolar lavage fluid, which was found to be correlated with increased collagen content and smooth muscle proliferation [21]. In an ovalbumin-induced asthma model featuring OPN knockout mice, a reduction in subcutaneous fibrosis, airway migration, and inflammatory cell infiltration was noted, along with diminished expression of TGF-β1 and VEGF, as well as decreased collagen deposition and smooth muscle actin synthesis, when compared to their wild-type counterparts [22, 23]. In CRSwNP patients, comparative analyses revealed a significant elevation in the protein levels of both OPN and VEGF compared to normal controls. It has been elucidated that OPN induces VEGF production in a manner that is contingent upon both time and dosage, mediated through the phosphatidylinositol 3-kinase-protein kinase B and extracellular signal-regulated kinase 1/2 signaling pathways [14]. In the context of our research, we observed a marked increase in the levels of OPN in nasal secretions among the CRSwNP cohort, with a further pronounced elevation in the eCRSwNP subgroup relative to the neCRSwNP subgroup.

OPN, in addition to its role in airway remodeling, is also closely related to eosinophil-mediated Th2-type inflammation. Eosinophil-derived OPN is posited to contribute to the restructuring of pulmonary tissue by engaging with VEGF, thereby facilitating neoangiogenesis [24]. It has been hypothesized that the administration of exogenous recombinant OPN can augment the pulmonary expression of IL-13 and matrix metalloproteinase-9, and is implicated in the thickening of the basement membrane through its binding to the ανβ3 integrin receptor, which subsequently activates the PI3K/Akt signaling pathway [22]. Furthermore, OPN derived from eosinophils has been shown to stimulate nasal fibroblasts, thereby promoting inflammation and tissue remodeling in eCRSwNP [25]. The expression levels of OPN have been found to be commensurate with the severity of eCRSwNP. Notably, an elevated expression of OPN was observed in patients with concomitant nasal polyps and asthma. Immunohistochemical analysis revealed positive staining for OPN in epithelial cells, submucosal glands, infiltrating cells, and the extracellular matrix. The transcription of OPN mRNA is modulated by various cytokines; it is induced by IFN-γ, IL-1β, and TNF-α, yet is suppressed by IL-4 and IL-13. Conversely, OPN has been demonstrated to induce the production of a spectrum of cytokines, including IFN-γ, IL-4, IL-5, IL-13, IL-1β, and TNF-α, within the sinus mucosa [26]. In addition, recombinant OPN significantly promoted eosinophil migration and enhanced eosinophil cationic protein (ECP) production in an in vitro dispersed NP cell culture system [25]. Our findings corroborate these observations, indicating that increased concentrations of OPN in nasal secretions are positively correlated with blood eosinophil counts and percentages, tissue eosinophil counts.

Currently, the overall therapeutic efficacy of CRS is deemed satisfactory; however, due to its pronounced heterogeneity and intricate pathophysiological mechanisms, the condition is characterized by a notably high rate of recurrence and resistance to treatment [1‒3]. A 12-year longitudinal study conducted by Bachert et al. reported an alarming recurrence rate of up to 80% for CRSwNP. This high recurrence rate not only significantly impairs the quality of life for patients with CRS, diminishes their capacity for learning and work, but also imposes a substantial economic burden on the healthcare system, underscoring the urgent need for more precise subtyping of CRS [27]. As our comprehension of the endotypes of CRS deepens and the development of biological immunotherapies advances, an array of randomized, double-blind, controlled trials has demonstrated that biological agents such as dupilumab, mepolizumab, or omalizumab exhibit enhanced efficacy in patients exhibiting Th2-mediated inflammation [28, 29]. Alsharif et al. [30] conducted “Reboot” surgery on CRS patients with pronounced Th2 inflammation, and subsequent follow-up data indicated a significant reduction in both the recurrence rate and the time to recurrence of the disease. This approach is instrumental in facilitating the accurate diagnosis and targeted treatment of CRS by tailoring treatment modalities and surgical interventions in accordance with the endotypes of CRS. Furthermore, our research has identified that OPN levels in nasal secretions serve as a more robust predictor of CRSwNP endotypes than traditional biomarkers such as blood eosinophil count, blood eosinophil percentage, serum total IgE, and FnNO. In addition, the classification of CRSwNP by OPN concentration was more accurate than that by eosinophil-dependent endotypes in VAS score, Lund-Mackay score, and Lund-Kennedy score, suggesting that OPN in nasal secretions may be a reliable biomarker for gauging the severity of CRSwNP.

This investigation is beset by several limitations that warrant resolution. Initially, the cohort of patients enrolled is comparatively diminutive, thereby attenuating the robustness of the conclusions. Subsequently, the study has exclusively enrolled Chinese patients who received treatment at a solitary medical facility, and their inflammatory profiles do not encapsulate the global demographic. Consequently, there is an exigency for subsequent studies with an expanded sample size and a more diverse ethnic spectrum across multiple centers to substantiate the findings of the present research.

In summary, OPN levels in nasal secretions are elevated in CRSwNP and exhibit a significant correlation with the extent of mucosal eosinophilic infiltration. The presence of OPN in nasal secretions emerges as a promising biomarker, offering a novel avenue for differentiating the endotypes of CRSwNP and for the prognostication of disease severity. The stratification of CRSwNP endotypes, based on this biomarker, lays the groundwork for the formulation of targeted therapeutic strategies tailored to individual patient profiles. This approach is anticipated to empower clinicians to mitigate the recurrence rate of CRSwNP and to enhance the postsurgical quality of life through personalized patient care.

This study involving human participants was approved by the Ethics Committee of Renmin Hospital of Wuhan University (No. WDRY2022-K227); all the participants were enrolled after writing informed consent.

The authors have no conflicts of interest to declare.

This work was supported by grants from the National Natural Science Foundation of China (NSFC): No. 82271134 (Yu Xu), No. 82401322 (Peiqiang Liu); the Natural Science Foundation of Hubei Province: No. 2023AFB229 (Peiqiang Liu); the Fundamental Research Funds for the Central Universities: No. 2042023kf0025 (Peiqiang Liu); and the Hubei key laboratory opening project: No. 2023KFZ011 (Peiqiang Liu).

P.L. and Y.X. oversaw the study, designed and conceived the study, prepared the figures, wrote the manuscript, and reviewed the manuscript for important intellectual content. P.L., M.L., Y.S., M.Y., and W.L. collected samples, performed experiments, and analyzed data. All authors read and approved the manuscript.

Peiqiang Liu, Meng Liu, and Yibin Sun contributed equally to this work.Edited by: D.Y. Wang, Singapore.

All data generated or analyzed during this study are included in this article. Further inquiries can be directed to the corresponding author.