The Efficacy of Dupilumab as an Adjuvant Treatment After Endoscopic Sinus Surgery for Chronic Rhinosinusitis With Nasal Polyps: A Retrospective Cohort Analysis

Abstract

Objectives.

Chronic rhinosinusitis with nasal polyps (CRSwNP) demonstrates a high recurrence rate following endoscopic sinus surgery (ESS). The role of adjunctive biologic therapy remains incompletely understood. This study aims to assess the efficacy of dupilumab when used in conjunction with ESS for patients with CRSwNP.

Methods.

We conducted a retrospective cohort study of type 2 CRSwNP patients who underwent ESS. The study compared patients receiving postoperative dupilumab with those receiving standard care. Changes in the Lund-Kennedy endoscopic score (LKES) and nasal polyp score (NPS) were recorded, and linear mixed models (LMMs) were employed for the primary outcome analysis.

Results.

The study enrolled 49 CRSwNP patients, with 22 patients in the dupilumab group. Postoperative LKES and NPS decreased in both groups. LMM analysis revealed a significant improvement in LKES (B=–0.61; 95% CI, –0.90 to –0.31; P<0.001) and a more pronounced improvement in NPS (B=–0.14; 95% CI, –0.25 to –0.02; P=0.025) in the dupilumab group compared to controls from preoperative to postoperative measurements.

Conclusion.

Compared to ESS alone, the adjunctive use of dupilumab in the postoperative period was associated with greater reductions in both NPS and endoscopic scores.

INTRODUCTION

The development of biologic agents targeting type 2 immune effectors—including interleukin (IL)-4, IL-13, IL-5, and immunoglobulin E (IgE)—has paved the way for a new generation of therapeutic options in managing severe chronic rhinosinusitis with nasal polyps (CRSwNP). Since 2019, several large-scale, randomized controlled trials have demonstrated the efficacy of biologic therapy in CRSwNP [1-4]. Following the European Forum for Research and Education in Allergy and Airway Diseases (EUFOREA) Consensus, biologics are now widely adopted for severe CRSwNP cases that meet at least three of the following criteria: evidence of type 2 inflammation, use of oral corticosteroids within the past two years, loss of smell, significantly impaired quality of life, and comorbid asthma [5,6]. Dupilumab, a monoclonal antibody that targets the IL-4 receptor alpha subunit and its downstream signaling via IL-4 and IL-13, has been shown to reduce polyp size and sinus opacification, improve olfaction and nasal symptoms, enhance asthma control, and lower the incidence of subsequent sinus surgery [1,7-10]. In both clinical trials and real-world settings, dupilumab has been administered to patients regardless of whether they have undergone surgery before [1,11-15].

Endoscopic sinus surgery (ESS) remains the primary treatment modality for CRSwNP [16]. Recent comparative studies have demonstrated that both ESS and dupilumab significantly improve nasal symptoms and quality of life [17-20]. Generally, ESS produces a greater reduction in polyp burden, whereas dupilumab yields superior improvements in olfaction-related symptom scores—especially in patients with a history of surgery or contraindications to anesthesia [17-21]. Furthermore, dupilumab has proven more effective than ESS in reducing the frequency of acute rhinosinusitis episodes and lowering the need for oral corticosteroids [22].

Given the diverse benefits offered by ESS and biologic therapies, it is worthwhile to explore their combined use in the management of CRSwNP. The potential of using biologics in conjunction with surgery has recently garnered increased attention, and combination therapy shows promise for achieving more substantial improvements [19]. ESS efficiently reduces the inflammatory burden and establishes drainage for improved ventilation and topical medication delivery. Concurrently, dupilumab’s efficacy in controlling mucosal inflammation suggests that its postoperative use may enhance epithelial recovery, decrease mucosal swelling, and expedite olfactory restoration. Although the combination may offer clinical and economic benefits for the treatment of CRSwNP, evidence for the use of ESS with adjunctive biologic therapy remains limited. Consequently, we conducted a retrospective cohort study comparing outcomes between patients receiving ESS with postoperative dupilumab and those undergoing ESS alone.

MATERIALS AND METHODS

This study was approved by the Institutional Review Board of National Taiwan University Hospital (No. 202310053RINB) and conducted in accordance with its ethical standards. Written informed consent was waived for the patients.

Study design and participants

This retrospective cohort study was conducted at a tertiary care center between 2020 and 2023. Adult patients diagnosed with type 2 CRSwNP who underwent ESS were enrolled. All patients had persistent sinonasal symptoms despite receiving at least three months of intranasal corticosteroid therapy prior to surgery. Type 2 inflammation was defined by preoperative assessments demonstrating either elevated serum IgE levels (≥100 IU/mL) or a blood eosinophil count ≥200 cells/μL. Patients were informed of the potential benefits and costs of dupilumab, and the decision to initiate postoperative treatment was made independently of study participation. Similarly, the decision to proceed with ESS was not influenced by study enrollment. In the dupilumab group, the medication was administered at a dose of 300 mg within 2 days following ESS, with a second dose given after 2 weeks, followed by monthly injections. All treatments were prescribed and managed by the same rhinologist. To be included in the dupilumab group, patients had to have received at least three doses. The control group consisted of patients diagnosed with type 2 CRSwNP before December 2021, prior to the national approval of dupilumab for CRSwNP, to minimize selection bias. Both groups underwent bilateral sinus surgeries and subsequently received standard postoperative care, which included twice-daily topical steroid irrigation and regular follow-ups. Patients were excluded if they had received monoclonal antibodies or immunosuppressive therapy within one month prior to screening, or had a history of immunodeficiency, coagulopathy, pregnancy, or cystic fibrosis. Patients with asthma or aspirin-exacerbated respiratory disease were not excluded, and these comorbidities did not alter the treatment strategy.

Based on previous studies comparing nasal polyp score (NPS) changes between ESS and dupilumab, the 6-month postoperative NPS change was –5 (standard deviation [SD], 1.75) in the ESS group and –3 (SD, 1.5) in the dupilumab group. A sample size of 19 per group was calculated to provide 80% power at an alpha level of 5% [20].

Clinical parameters

The following demographic information was collected: a brief history of rhinosinusitis symptoms and previous treatment, the presence of concurrent asthma, past surgical history, smoking status, and allergies to medication. A serum-specific allergen test was performed utilizing the ImmunoCAP (Thermo Fisher) test to examine levels of specific IgE towards house dust mites, tree pollens, grass pollens, weed pollens, mold species, cat hair, dog hair, and cockroaches. The percentages of blood eosinophils and eosinophil counts in surgical specimens were obtained to assess eosinophilic inflammation. Tissue eosinophilia was diagnosed based on an eosinophil count of 10 or more per high-power field, in accordance with the 2020 European Position Paper on Rhinosinusitis and Nasal Polyps (EPOS) guidelines [23]. Sinonasal symptoms were evaluated by the Chinese version of the Sinonasal Outcome Test-22 (SNOT-22) before the operation. The SNOT-22 was scored on a scale of 0 to 5, with a total score ranging from 0 to 110. Higher scores indicate more severe symptoms. Lund-Mackay computed tomography (CT) scores evaluate the patency of each sinus using a 0 to 2 scale (0=normal, 1=partial opacification, 2=total opacification) and have a total score ranging from 0 to 24. The individual scores of distinct sinuses from the Lund-Mackay CT scoring system were employed in the calculation of the ethmoid/maxillary ratio. This ratio was defined as the average of the anterior ethmoid and posterior ethmoid scores divided by the maxillary score. In the event that both maxillary sinus scores were zero, the ethmoid/maxillary ratio was defined as the average of the ethmoid scores. The TIB (Top International Biotech) Smell Identification Test (TIBSIT) was employed as a bilateral psychophysical olfactory function test. TIBSIT is an array of 16 questions, wherein subjects scratch the odor strip, sniff, and complete the questionnaire. The TIBSIT scores range from 0 to 48, with scores of 45–48, 15–44, and 2–12 corresponding to normosmia, hyposmia, and anosmia, respectively [24].

Outcome measurements

The preoperative Lund-Kennedy Endoscopic Score (LKES) and NPS were documented at the operating theater prior to surgery [25,26]. Postoperative visits were categorized into intervals of 2 weeks, 6 weeks, 3 months, and 6 months. The data were collected from the visit that occurred closest to the end of each interval. During the follow-up visits, the LKES and NPS were recorded in order to evaluate the postoperative outcomes. The SNOT-22 questionnaire was administered at the 6-month postoperative interval.

Statistical analysis

A comparison was conducted between the baseline characteristics of patients who received ESS and postoperative dupilumab, and those who underwent surgery alone. This was done using Fisher’s exact test for categorical variables, independent sample t-test for continuous variables, and Mann-Whitney U-test for obviously skewed continuous variables (e.g., IgE). The changes in LKES and NPS from preoperation to post-operation measurements (i.e., 2nd week, 6th week, 3rd month, and 6th month) between the study groups were compared using a linear mixed model. The linear mixed model incorporated the main effects of the study group (dupilumab vs. control) and time (expressed as month) and an interaction term between group and time. The linear mixed model included a random intercept to account for repeated measurements per subject. To investigate potential factors associated with LKES and NPS at postoperative events in the adjuvant dupilumab group, a series of univariate LMMs were conducted. All tests were conducted in a two-tailed manner, and a P-value of less than 0.05 was considered statistically significant. The data were analyzed using IBM SPSS version 26 (IBM Corp.). The plot was generated using GraphPad Prism 9.0 software (GraphPad Software).

RESULTS

Clinical characteristics in the cohort

After applying inclusion and exclusion criteria (Fig. 1), 49 CRSwNP patients were included in the study. Between January 2022 and August 2023, 143 CRSwNP patients underwent ESS; among these, 84 met the criteria for type 2 CRS based on surgical specimens. Postoperatively, 27 patients (32.1%) received dupilumab, and five patients who received fewer than three doses were excluded from further analysis. In a previous period from June 2020 to December 2021, 130 CRS patients underwent ESS, with 53 meeting the criteria for type 2 CRS; of these, 27 patients who completed at least three postoperative visits were included in the control group. Among the 49 patients analyzed, 47 procedures were performed by one surgeon and 2 by another.

Fig. 1.

Flowchart of patient selection. ESS, endoscopic sinus surgery; IgE, immunoglobulin E.

A comparison of the demographic characteristics between the groups (Table 1) revealed no significant differences in age, sex, asthma status, smoking history, allergen sensitization, history of prior ESS, olfactory function, or the performance of concurrent septoplasty. With regard to type 2 inflammation, IgE levels were similar between groups, although eosinophil counts were significantly higher in the dupilumab group (median [interquartile range], 570 [398–726] vs. 271 [217–443]; P=0.003). The preoperative Lund-Mackay score and ethmoid/maxillary ratio on CT were not significantly different between the groups, nor were the preoperative LKES and NPS. All patients in this study exhibited pathologically confirmed tissue eosinophilia.

Table 1.

Baseline characteristics of the patients in the dupilumab and control groups

Variable	Available number	Total (n=49)	Adjuvant dupilumab group (n=22)	Control group (n=27)	P-valuea)
Age (yr)	49	50.3±11.3	50.1±8.6	50.4±13.3	0.915
Sex, male	49	32 (65.3)	13 (59.1)	19 (70.4)	0.548
Doses of dupilumab injected	22	4.2±1.0	4.2±1.0	-	-
Asthma	49	23 (46.9)	13 (59.1)	10 (37.0)	0.157
Smoking	49	5 (10.2)	1 (4.5)	4 (14.8)	0.362
Revised surgery	49	14 (28.6)	9 (40.9)	5 (18.5)	0.116
Blood eosinophil count (/μL)	49	417 (241–607)	570 (398–726)	271 (217–443)	0.003
IgE (IU/mL)	25	179 (91–453)	266 (112–453)	136 (65–477)	0.628
Tissue eosinophilia	49	49 (100)	22 (100)	27 (100)	1.000
Allergen sensitization	39	22 (56.4)	9 (64.3)	13 (52.0)	0.518
Ethmoid/maxillary ratiob)		1.24±0.40	1.27±0.45	1.22±0.37	0.700
Olfactory function	38	16.0±16.7	13.6±16.2	18.0±17.3	0.432
Normal		4 (10.5)	2 (11.8)	2 (9.5)	1.000
Hyposmia		9 (23.7)	3 (17.6)	6 (28.6)	0.476
Anosmia		25 (65.8)	12 (70.6)	13 (61.9)	0.734
Lund-Mackay score	49	15.3±4.5	16.1±4.5	14.7±4.5	0.286
SNOT-22 scores	46	51.2±26.8	48.5±28.6	53.3±25.7	0.545
Preoperative LKES	49	8.0±2.7	7.6±3.3	8.3±2.1	0.401
Preoperative polyp score	49	4.3±1.7	4.5±1.7	4.3±1.8	0.696
Concurrent septoplasty	49	17 (34.7)	10 (45.5)	7 (25.9)	0.228
Follow-up duration (mo)	49	4.6 (3.2–6.3)	4.6 (3.1–6.3)	4.2 (3.2–6.7)	0.732

Values are presented as mean±standard deviation, number (%), or median (interquartile range).

IgE, immunoglobulin E; SNOT-22, Sinonasal Outcome Test-22; LKES, Lund-Kennedy endoscopic score.

^a)Comparison between groups using the Fisher exact test for categorical variables, the independent sample t-test for continuous variables, and the Mann-Whitney U-test for obviously skewed continuous variables.

^b)Ethmoid/maxillary ratio: average of the anterior ethmoid and posterior ethmoid Lund-Mackay scores divided by the maxillary Lund-Mackay score.

Changes in endoscopic findings and SNOT-22

In the dupilumab group, 20 out of 22 patients completed four visits within the designated timeframe, while 23 out of 27 patients in the control group completed four visits up to 6 months postoperatively. The LMM analysis demonstrated that the dupilumab group experienced a significant improvement in LKES from preoperative to postoperative measurements (B=–0.61; 95% CI, –0.90 to –0.31; P<0.001) (Fig. 2A). Similarly, the improvement in NPS was more pronounced in the dupilumab group (B=–0.14; 95% CI, –0.25 to –0.02; P=0.025) (Fig. 2B).

Fig. 2.

The mean values of Lund-Kennedy endoscopic score (LKES; A) and nasal polyp score (NPS; B) of patients at all measurement occasions in the dupilumab and control groups. The error bar represents the standard error of mean.

Both groups showed significant improvement in SNOT-22 scores postoperatively. In the dupilumab group, the mean score changed from 53.3±25.7 at baseline to 14.7±12.7 (P<0.001), and in the control group from 48.5±28.6 to 12.2±16.4 (P<0.001). No statistically significant difference in the 6-month postoperative SNOT-22 scores was noted between the groups (control: 14.7±12.7; dupilumab: 12.2±16.4; P=0.577) (Table 2).

Table 2.

SNOT-22 score before and after 6 months of treatment in the dupilumab and control groups

Variable	Before surgery	6 Months after surgery	P-valuea)
Total	51.2±26.8	13.5±14.4	<0.001
Adjuvant dupilumab group	48.5±28.6	12.2±16.4	<0.001
Control group	53.3±25.7	14.7±12.7	<0.001
Pairwise P	0.545	0.577	-

Values are presented as mean±standard deviation.

SNOT-22, Sinonasal Outcome Test-22.

^a)Comparison between groups using the independent sample t-test.

Effect modifiers

Univariate LMMs were employed to identify factors associated with changes in LKES and NPS following ESS in the adjuvant dupilumab group (Fig. 3). A higher preoperative IgE level was associated with higher postoperative LKES (B=0.12; 95% CI, 0.06 to 0.18). Smoking was associated with higher postoperative NPS (B=0.35; 95% CI, 0.16 to 0.55) while concurrent septoplasty was associated with lower postoperative NPS (B=–0.37; 95% CI, –0.71 to –0.03). No significant association was observed between the number of dupilumab doses and postoperative LKES (B=0.05; 95% CI, –0.58 to 0.68) or total NPS (B=0.05; 95% CI, –0.14 to 0.23).

Fig. 3.

Univariate linear mixed-effects model analysis of the association between baseline characteristics and postoperative Lund-Kennedy endoscopic score (LKES; A) and nasal polyp score (NPS; B) among patients receiving adjuvant dupilumab. IgE, immunoglobulin E; CT, computed tomography.

DISCUSSION

Biologics have been used to treat type 2 CRSwNP; however, the efficacy of adjunctive dupilumab in the postoperative period remains not fully elucidated. Moreover, short-term outcomes of dupilumab use have been reported in only a limited number of studies. In this retrospective cohort study comparing ESS with and without postoperative dupilumab, we demonstrated significant improvements in both LKES and NPS in the dupilumab group.

Surgery plays a critical role in managing CRSwNP, often resulting in a greater reduction in nasal polyp burden than dupilumab alone [17-21]. In one multicenter study comparing CRSwNP patients undergoing ESS with phase 3 trial data of biologics, ESS produced significantly lower NPS at 24 and 52 weeks postoperatively compared to dupilumab, omalizumab, and mepolizumab [17]. Other retrospective studies have also observed a greater reduction in NPS in the ESS group compared to the dupilumab group [17,18,20]. Interestingly, De Corso et al. [18] reported that although the ESS group experienced a significant initial decrease in NPS, the scores gradually increased from 3 to 12 months postoperatively, indicating that ESS provided superior NPS reduction for up to six months, after which no difference was noted between the two groups. Furthermore, in patients with aspirin-exacerbated respiratory disease, no significant difference in NPS reduction was observed between surgery and dupilumab treatment [17]. Overall, ESS offers an efficient and substantial reduction in nasal polyp burden, but in patients with a pronounced inflammatory process, the differences between surgery and dupilumab may be less evident.

Our findings indicate that adding dupilumab to ESS results in superior improvements in LKES and NPS compared to surgery alone. Although previous studies have shown better NPS outcomes with surgery compared to dupilumab alone, our combination approach—employing early surgery with adjunctive biologics—may benefit patients with severe CRSwNP. The linear mixed model produced B values of –0.61 for LKES and –0.14 for NPS. Extrapolated over six months, this suggests that the adjunctive dupilumab group may achieve an LKES reduction of approximately 3.6 points and an NPS reduction of 0.84 points compared to the control group. However, no significant difference in SNOT-22 scores was observed at six months, indicating that further research is needed to determine whether these endoscopic improvements correlate with long-term symptom relief. Notably, 10 of the 22 patients in the dupilumab group underwent concurrent septoplasty and exhibited lower postoperative NPS. This may reflect enhanced surgical access or improved postoperative care and aligns with previous studies reporting lower revision ESS rates when septoplasty is performed concurrently [26-28]. Additionally, a higher preoperative IgE level was associated with higher postoperative LKES, potentially due to a greater systemic inflammatory burden; however, it remains unclear whether this necessitates a higher dupilumab dose. Given the retrospective design, certain limitations must be acknowledged. For example, the dupilumab group had significantly higher baseline blood eosinophil counts, suggesting a more severe disease phenotype and potentially contributing to modest short-term improvements. Furthermore, surgeons may be inclined to preserve the olfactory epithelium in patients known to receive postoperative dupilumab, which could lead to transient postoperative mucosal swelling. This study is intended as a pilot for future investigations into the combined use of ESS and biologics, Boptimal biologic dosing, and the identification of potential risk factors for suboptimal responses.

In a post-hoc analysis based on sinus-24 and sinus-52 data by Hopkins et al. [29], improvements in NPS and Lund-Mackay score were greater in patients whose last surgery occurred less than three years prior. A plausible explanation is that a shorter interval since the previous surgery results in a lower inflammatory load in the sinus tissue, thereby enhancing the effect of dupilumab. Consequently, we initiated dupilumab immediately after surgery to intervene before sinus remodeling commenced. Furthermore, after the third dose, we shifted to monthly dupilumab administration rather than every 2 weeks, based on the hypothesis that early intervention combined with the effects of ESS might allow for a lower dupilumab dose compared to using dupilumab alone. A retrospective case-control study by Garvey et al. [19] found that the combination of ESS and biologic therapy resulted in the greatest reduction of sustained polyp burden compared to biologic therapy alone. This combined approach not only alleviates polyp burden but also reduces inflammation, thereby enhancing medication delivery and overall care, which may maximize and prolong treatment benefits.

In a recent report, Wu et al. [30] investigated the efficacy of combining biologics with ESS in a cohort of 10 patients receiving postoperative dupilumab and 20 control subjects. At 3 months postoperatively, the dupilumab group exhibited superior SNOT-22 improvements and better LKES [31]. In the present study, the change in SNOT-22 scores from preoperative to postoperative assessments demonstrated significant improvement. Although significant improvements in endoscopic scores were observed in the combined surgery and dupilumab group, no significant difference was noted in SNOT-22 scores between the dupilumab group and the control group at 6 months postoperatively. It is noteworthy that discrepancies between subjective and objective evaluations in CRS patients may exist. Therefore, further investigation into the postoperative effects of biologics is warranted in a larger sample with a longer observation period.

A preference for biologic treatment is evident among CRS patients, particularly in those with comorbid asthma or non-steroidal anti-inflammatory drugs-exacerbated respiratory disease [31,32]. However, concerns about the high cost of biologics persist, especially in the absence of health insurance coverage. Previous studies have reported that the mean cost of uncomplicated ESS ranges from $7,500 to $9,000 [33,34], while the annual cost of dupilumab is approximately $31,000 [35]. Cost-utility analyses by Scangas et al. [36] and Parasher et al. [37] consistently identified ESS as the more cost-effective option compared to dupilumab. Moreover, in many countries, dupilumab is not covered by health insurance, further increasing patients’ economic burden. Extending the interval between dupilumab injections represents one potential avenue for reducing treatment costs. Additionally, the favorable outcomes observed in our study regarding nasal polyp control suggest that the need for revision surgeries may be diminished, thereby contributing to overall cost savings and improved patient outcomes.

The key strength of this study lies in its standardized postoperative dupilumab regimen and detailed endoscopic follow-up. By including a control group that received ESS without dupilumab, the study provides a comparative analysis of the additional benefits of dupilumab in postoperative management. Furthermore, the application of linear mixed models to analyze changes in LKES and polyp scores over time allowed for appropriate handling of repeated measurements and adjustment for potential confounders. However, several limitations should be acknowledged. First, the sample size is relatively small, and the study was conducted retrospectively. To minimize selection bias, we selected control subjects from a period before dupilumab became available. Second, due to the retrospective design, some data—particularly symptom scores and olfactory tests—were missing, preventing us from correlating these outcomes with the endoscopic findings. Third, the study population consisted solely of Asians, who may differ in CRSwNP characteristics from other populations, as previous studies have reported.

The efficacy of dupilumab as an add-on treatment remains uncertain, underscoring the need for a prospective study comparing ESS alone, dupilumab alone, and their combination. Additionally, investigating the optimal frequency of dupilumab administration as an adjunct therapy may help establish a more economically sustainable treatment strategy. Further research is required to determine the optimal dose and duration of dupilumab therapy for CRSwNP patients. Moreover, although patients generally prefer dupilumab to repeated surgeries, the appropriate extent of surgery necessary before initiating dupilumab treatment remains unclear. Long-term observations are essential to address these issues and refine treatment protocols for CRSwNP.

In comparison to ESS alone, the adjunctive use of dupilumab in the postoperative period has been shown to produce more substantial reductions in both NPS and endoscopic scores. Higher preoperative IgE levels were associated with increased postoperative LKES, while smoking was linked to higher postoperative NPS, and concurrent septoplasty correlated with lower postoperative NPS. Given the retrospective nature of this study, a definitive causal relationship cannot be established. Nevertheless, a longer observation period may yield further insights into the potential role of postoperative dupilumab in reducing the recurrence of nasal polyps and the need for revision surgery or ongoing biologic therapy.

HIGHLIGHTS

▪ This study reflects the real-world use of dupilumab in a setting where access is limited due to reimbursement restrictions and the high cost of biologics for chronic rhinosinusitis with nasal polyps (CRSwNP) treatment.

▪ Early initiation of dupilumab after endoscopic sinus surgery significantly improved endoscopic scores (Lund-Kennedy Endoscopic Score) and nasal polyp scores, suggesting superior control of local inflammation and reduced polyp burden in patients with type 2 CRSwNP.

▪ Multiple postoperative evaluations and linear mixed models were used to assess the treatment effect of dupilumab over time.

▪ Continued longitudinal follow-up will determine whether postoperative dupilumab can influence the chronic inflammatory trajectory of CRSwNP and contribute to durable disease control.

▪ Further investigation into the optimal timing, dosing, and duration of dupilumab is needed to develop economically sustainable treatment strategies and optimize treatment costs.