Utility of Nasal Mucus Inflammatory Profile as a Biomarker of Nasal Polyp Regrowth in Aspirin-Exacerbated Respiratory Disease

Aspirin-exacerbated respiratory disease (AERD) is a disease of the upper and lower airways characterized by asthma, chronic rhinosinusitis with nasal polyposis (CRSwNP), and acute upper and/or lower airway symptoms triggered by cyclooxygenase-1 inhibitors. Individuals with AERD have more severe upper and lower airway disease, poorer quality of life (QOL), and increased medication utilization when compared to their aspirin-tolerant asthma and CRSwNP counterparts(1). The high rate of nasal polyp recurrence despite adequate surgical excision in patients with AERD is major driver of health care utilization and impaired QOL(2). No validated disease progression or severity biomarkers exist for AERD.

Traditionally, AERD is associated with robust type (T)2 eosinophilic inflammation; recent data supports inflammatory heterogeneity in both the upper(3) and lower(4) airways. We recently reported three distinct inflammatory clusters based on hierarchical cluster analysis of nasal mucus cytokines(3). Cluster 1 (33%) was characterized by relative low inflammation. Cluster 2 (47%) demonstrated high T2 markers, accompanied by high interleukin-6. Cluster 3 (20%) was marked by a mixture of T1 and T3 inflammation (Table E1). The impact of inflammatory endotype on post-operative clinical outcomes was not investigated. We hypothesized that the nasal mucus inflammatory profile at the time of surgery would impact time-to-polyp regrowth, time-to-systemic steroid course, and initiation of biologic therapy.

Scott et al previously identified three unique inflammatory clusters in subjects with AERD by hierarchical clustering of intraoperative nasal mucus cytokine levels (3). Retrospective chart review of the original AERD cohort (n=30, defined by history of asthma, nasal polyps, and two respiratory reactions with COX-1 ingestion or positive challenge), enrolled between 2015 and 2019, captured routine clinical care in the otolaryngology and allergy clinics to assess for nasal polyp regrowth, disease-related medical interventions, and QOL. We excluded patients with less than 12 months of post-operative otolaryngology follow-up and ended data collection at 24 months post-functional endoscopic sinus surgery (FESS). Time at which polyp regrowth was noted on nasal endoscopy, time at which first post-operative systemic steroid for asthma exacerbation or nasal polyposis was prescribed, and initiation or change in prescription of a biologic approved for asthma and/or CRSwNP were recorded. Kruskal-Wallis or Fisher’s Exact test was used to identify differences in demographic, baseline clinical information and biologic utilization among cluster groups (Table E1). Kaplan-Meier survival analysis and log-rank test identified differences in time-to-polyp regrowth and time-to-systemic steroid by cluster with Cox proportional hazards regression performed for individual comparisons.

Twenty patients met inclusion criteria: six in cluster 1, ten in cluster 2, and four in cluster 3. Ten patients were excluded due to inadequate nasal endoscopy data after FESS. There were no differences in demographic and baseline clinical characteristics between the original cohort and those meeting study criteria (Table E2). No differences in demographic and baseline clinical characteristics were observed among the three inflammatory clusters (Table E1). It is noted that all individuals in cluster 3 are male and a smaller proportion of patients in cluster 2 were started on aspirin therapy, though the sample size was not large enough to detect a statistical significance. All individuals were prescribed topical intranasal budesonide after surgical intervention; one individual (cluster 1) was known to be non-compliant.

Time-to-polyp regrowth was significantly different among cluster groups (p=0.03, Figure 1). Investigative regression analysis demonstrated that subjects in cluster 3 were more likely to have polyp regrowth by the end of the study period compared with clusters 1 (HR 7.5, 95% CI 1.2-47.4) and 2 (HR 7.5, 95% CI 1.2-44.9). It is important to note that all four subjects in cluster 3 demonstrated regrowth of polyps within 14 months of FESS. Biologic utilization differed among clusters (p<0.05). Subjects in cluster 3 were more likely to initiate biologics compared to cluster 1 (p<0.05) with a similar trend observed when compared to cluster 2 (p=0.07); all subjects in cluster 3 initiated a new biologic therapy during the study period (Table E1). A trend was observed in time-to-systemic steroid use among the three clusters (p=0.09, Figure 2). QOL data was too limited for analysis.

Figure 1.

Time-to-Polyp Regrowth curves by inflammatory group. Polyp regrowth defined by frank polyposis on nasal endoscopy. Censored individuals (◆) correspond to those without clinical data to 24 months and no event at last known clinical encounter. Individuals on aspirin therapy at polyp recurrence or censoring indicated by a plus (+). Asterisk (*) indicates p<0.05.

Figure 2.

Time-to-Steroid curves by inflammatory group. Steroid defined by first systemic steroid prescription after surgery, not inclusive of immediate post-operative steroid courses. Censored individuals (◆) correspond to those without clinical data to 24 months and no event at last known clinical encounter. Individuals on aspirin therapy at polyp recurrence or censoring indicated by a plus (+).

These results support that inflammatory heterogeneity observed in AERD has clinical implications. Subjects with a lower relative inflammatory burden (cluster 1) demonstrated the lowest clinical disease burden. While AERD is classically associated with a high T2 inflammatory burden, subjects with a mixed T1 and T3 cytokine profile (cluster 3) have the most refractory clinical disease with shorter time-to-polyp regrowth and greater need for biologic therapy. These findings are supported by previous studies that have demonstrated decreased QOL and more revision surgeries in those with a T3 inflammatory endotype(5, 6).

Medical interventions for nasal polyposis in AERD are few; sinus steroid irrigations, aspirin therapy after desensitization and biologics are the most effective options(2, 7). There are significant costs associated with revision FESS in those who fail medical treatments and biologics, totaling up to $40,000 per year(2, 7). Therefore, it is important to understand which patients are most likely to require additional interventions to guide management and healthcare utilization. All individuals in cluster 3 initiated and demonstrated clinical response to dupilumab despite the T1/T3 profile. Recall that individuals in cluster 3 possess T2 inflammation, just to a lesser degree. Our data support that inflammatory profile may inform post-FESS prognosis, but not necessarily response, to targeted biologic agents.

While this study provides important insights, the modest sample size, the heterogenous use of aspirin therapy, and the retrospective study design are limitations. Though significant and clinically relevant differences in time-to-polyp regrowth and biologic initiation by inflammatory cluster were observed, a prospective study and larger cohort may reveal additional clinical differences and narrow the wide confidence interval of Hazard Ratio analysis. The lack of sufficient data from patient-reported outcome measures limits our ability to assess the impact of nasal mucus inflammatory profile on QOL.

In 20 subjects with AERD, nasal mucus inflammatory profile at time of surgery influenced post-surgical time-to-polyp regrowth and biologic initiation. Prospective studies assessing the utility of nasal mucus cytokine profiles as a biomarker of disease severity, nasal polyp regrowth and need for biologic therapy are needed in CRSwNP and AERD.

Clinical Implications.

This study supports that inflammatory endotype impacts post-endoscopic sinus surgery clinical outcomes in patients with AERD. Nasal mucus inflammatory profiling may inform chronic rhinosinusitis with nasal polyposis progression and severity in AERD.