Dupilumab Effectiveness in Biologic‐Naïve and Switched Severe Asthma in Real Life

This retrospective observational study reports our 18-month real-life experience with dupilumab in 115 patients with severe asthma (including 68 patients who switched from another biologic: 17 from anti-IgE, 32 from anti-IL-5, and 19 from anti-IL-5R, and 47 biologic-naïve patients) recruited at the University Hospital of Liege, Belgium between 2020 and 2025. The time between the switch from the prior biologic to dupilumab was 44 (35–52) days and the reasons of the change were: persistent exacerbations (n = 10); uncontrolled nasal polyposis (n = 15); concomitant severe atopic dermatitis (n = 3); lack of asthma control (n = 5); or a mix of those associated with fall of lung function and/or inability to stop maintenance oral corticosteroids (mOCS, n = 35) Out of the 115 patients, 64 had an 18-month visit. Only 13 stopped the dupilumab after 6 months due to insufficient response, mainly in terms of exacerbations persistence or absence of nasal polyposis/asthma control. Recent randomized clinical trials (RCTs) showed that dupilumab improved asthma control, reduced the exacerbation rate and fractional exhaled nitric oxide (FeNO), and improved lung function and forced oscillation technique (FOT) parameters in patients with moderate-to-severe asthma [1, 2]. In this study, we analyzed, as primary outcome, the impact of dupilumab on exacerbation rate. The secondary outcomes included the impact on mOCS, daily dose of inhaled corticosteroids (ICS), asthma control and quality of life (assessed by ACT, ACQ, and AQLQ), lung function assessed by spirometry and FOT, systemic inflammation, and airway inflammation evaluated with FeNO and induced sputum cell counts. The demographic characteristics of the patients are presented in the Table S1 and the results were addressed for the whole cohort (Table S2), for the cohort of patients who switched, and for the biologic-naïve cohort (Tables S3 and S4). The evolution of exacerbation rate since baseline was studied using a negative binomial regression model with repeated measurements. The exacerbation rate significantly decreased from baseline (OR = 8.86, CI: 5.3–14.7, p < 0.0001 for 6 months vs. baseline and OR = 7.37, CI: 4.4–12.4, p < 0.0001 for 18 months vs. baseline). After 6 months, the average reduction was 88.8% ± 31.8% and was 91.0% ± 20.2% after 18 months in the whole population. The exacerbation rate was significantly reduced over time in the group of patients who switched and in the biologic-naïve group. The reduction was significantly higher in the group of biologic-naïve patients compared to patients who switched (adjusted OR = 8.76, CI: 5.3–14.6, p < 0.0001 for 6 months vs. baseline and adjusted OR = 7.45, CI: 4.2–13.1, p < 0.0001 for 18 months vs. baseline; group: adjusted OR = 3.03, CI: 2.0–4.6, p < 0.0001 between both groups, Figure 1A). The result was similar when the model was adjusted for age, smoking status (pack-years), presence of nasal polyposis, OCS and blood eosinophils. The linear mixed model was not able to assess the impact on OCS dose reduction due to a limited amount of data. The reduction of ICS dose equivalent was not significant. ACT, ACQ, and AQLQ scores significantly improved from baseline (p < 0.0001). All the lung function parameters evaluated positively over time. The pre-bronchodilation (BD) and post-BD FEV1 varied over time (p = 0.0007 or 0.0005 in L, p < 0.0001 in percentages), especially during the first 6 months after baseline with, on average, a lung function improvement (increase of 0.12 ± 0.36 L or 4.51% ± 11.57% in pre-BD and increase of 0.12 ± 0.34 L or 4.31% ± 10.90% in post-BD). When we looked closely, 58 patients (50.4%) and 27 patients (42.2%) had an increase of pre-BD FEV1 of at least 100 mL after 6 months and 18 months respectively. No significant evolution from baseline was detected for oscillometry. FeNO decreased significantly from baseline (p < 0.0001, decrease from 33 ppb at baseline to 19.5 ppb after 6 months or 16 ppb after 18 months). A significant change from baseline was detected for total serum IgE (p < 0.0001) and blood eosinophil counts (BEC, p < 0.0001). By adjusting for covariates, most continuous measures did not significantly differ between the switched and the biologic-naïve patients. Nevertheless, the ACQ score improved more significantly in the biologic-naïve group (interaction p = 0.035, Figure 1B), the BEC significantly increased from baseline in the switch group but not in the biologic-naïve group (interaction p < 0.0001, Figure 1C) and there was a significant decrease in plasma fibrinogen between baseline and the 18-month visit only for the switch group (interaction p = 0.0012, Figure 1D). A trend (interaction p = 0.051) for a higher decrease in sputum eosinophil percentage (SEP) was also observed in the biologic-naïve group after the first 6 months of dupilumab (Figure 1E). The rebound in BEC within the biologic-switched cohort was mainly observed in the group of patients treated with anti-IL5/IL-5R treatment (Figure 2 and Table S5), and a rise in BEC after stopping anti-IL-5 therapy was already shown in the literature [3]. In contrast, the SEP trajectory tended to diverge in the biologic-naïve group. Indeed, a trend for a decrease was observed in the biologic-naïve group that could be explained by a reduction in eotaxin production mediated by IL-4 and IL-13 [4]. The clinical significance of this increase in BEC was analyzed in the 11 patients who presented a BEC ≥ 1000 cells/ μl at 6 months and who had a follow-up at 18 months vs. 52 patients who presented a value < 1000 cells/ μl at 6 months and who had a 18-month visit. None had deleterious effects of this hypereosinophilia. This confirmed what was recently reported [4, 5] and showed that dupilumab, in contrast to anti IL-5 and anti-IL-5R therapy, does not control systemic eosinophilic inflammation but probably inhibits eosinophil recruitment and activation, thereby resulting in a clinical improvement. Univariate logistic regression was considered to model the association between an improvement in post-BD FEV1 expressed as predicted % (delta between 6 and 18 months ≥ 0% vs. < 0%) and a set of covariates. We found that the presence of nasal polyposis (OR:3; CI: 1.1–8.6, p = 0.04) and a previous treatment with mepolizumab (OR:3; CI: 1.048–9.276) were linked to a stable/improved lung function. We evaluated if a persistent high BEC at 6 months ≥ 1000 cells/ μl (OR = 8.1; CI: 1.8–56.5, p = 0.012) and a high SEP at 6 months (OR = 1.1; CI: 1.022–1;221, p = 0.026 for n = 37 data) impacted lung function and found that they were associated with a stable/improved lung function, suggesting the absence of eosinophil activation under dupilumab treatment. The multivariate logistic regression analysis combining mepolizumab and BEC at 6 months ≥ 1000 cells/ μl showed that the impact of mepolizumab disappeared when associated with high BEC. The model did not find significant association between BEC or SEP at 6 months after dupilumab and an improvement of ACT or ACQ at 18 months. Superresponse at 6 months and remission at 18 months were defined as 0 exacerbation, no OCS, ACT ≥ 20 or ACQ ≤ 1.5 and FEV1 ≥ 80%, we observed that 31% of patients (n = 36/115) were super-responders after 6 months with 23% of biologic-naïve patients (=11/47) and 37% of switchers (25/68). After 18 months of treatment with dupilumab, 41% patients achieved remission (=26/64) including 26% of biologic-naïve patients (=7/23) and 46% of patients who switched (=19/41). In conclusion, this study confirms the long-term efficacy of dupilumab in both naïve and biologic-switched patients [6]. Although the sample size was limited, this real-world analysis provides valuable insights into the multifaceted effects of dupilumab in biologic-naïve and switched severe asthma patients, a facet less explored in previous RCTs. Catherine Moermans and Mare Sabbe participated in the study design, performed the research and data analysis, and wrote the manuscript; Adrien Onssels, Noémie Bricmont, Romane Bonhiver, France Regnier, Sophie Graff, Sara Gerday, Makon-Sébastien Njock, Adeline Rosu, Virginie Paulus, Françoise Guissard, Stéphanie Ziant and Carole Sanchez participated in the sputum induction and processing and patient data collection. Marie Ernst performed and reviewed the statistical analyses. Christophe Desmet, Renaud Louis and Florence Schleich designed the study and interpreted the data. All authors participated in the manuscript reviewing and gave final approval of the manuscript and ensured that questions related to the accuracy or integrity of any part of the work were appropriately investigated and resolved. The authors thank the pneumology department staff of the CHU of Liege, Belgium, and the EOS program number O-0001422RG3330. The authors have nothing to report. Data Sharing: The datasets used during the current study are available from the corresponding author on reasonable request. Catherine Moermans, Mare Sabbe, Adrien Onssels, Noémie Bricmont, Romane Bonhiver, France Regnier, Sophie Graff, Sara Gerday, Makon-Sébastien Njock, Adeline Rosu, Virginie Paulus, Françoise Guissard, Stéphanie Ziant, Carole Sanchez, Marie Ernst and Christophe Desmet have no conflicts to declare. Renaud Louis and Florence Schleich had educational and research grants from GSK, AstraZeneca, Sanofi, Novartis, and Chiesi. Also, Renaud Louis and Florence Schleich received consulting fees from GSK and AstraZeneca (national and international advisory boards). Renaud Louis and Florence Schleich received lecture fees from GSK, AstraZeneca, Novartis, and Chiesi. The data that support the findings of this study are available from the corresponding author upon reasonable request. Table S1: Baseline characteristics of patients from the whole cohort (N = 115) and comparison between groups Table S2: Evolution of the clinical and inflammatory characteristics of the patients 6 and 18 months after dupilumab (n = 115) Table S3: Evolution of the clinical and inflammatory characteristics of patients with switch (n = 68) and for biologic-naïve patients (n = 47) 6 and 18 months after dupilumab. Table S4: Joint effect of time and groups on all continuous parameters adjusted by covariates (age, pack-years, presence of nasal polyposis, baseline OCS and baseline blood eosinophils) Table S5: Joint effect of time and previous biotherapy on blood eosinophils (/μL—log-transformed). Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.