Short report

Severe asthma is described by the ERS/ATS guidelines as asthma that requires treatment with high-dose ICS plus a second controller with or without systemic corticosteroids to maintain control of the disease or, that remains uncontrolled despite this therapy. Severe eosinophilic asthma represents a clinical inflammatory phenotype associated with greatest risks of exacerbation, hospitalizations, poor asthma control, comorbid nasal polyposis, impaired lung function, and is characterised by a significant number of sputum or blood eosinophils.1 IL-5 is the key cytokine involved in the maturation, activation and survival of eosinophils and anti–IL-5 biologics were developed to reduce eosinophilic inflammation by inhibiting this type-2 pathway. Among these treatments, mepolizumab is a mAb that targets the ligand IL-5 and benralizumab depletes eosinophils by binding to the IL-5 receptor.2 Several clinical trials have demonstrated that anti-IL-5(R) biologics were able to improve lung function, asthma control and chronic oral corticosteroid (OCS) exposure and reduce exacerbations among eosinophilic asthmatic patients.3 4 Some recent studies have however brought to light a certain variability in clinical responses to anti-IL-5(R) biologics.5–7 Indeed, it appears that some patients are able to reach complete asthma control and some experience sub-optimal response with residual disease manifestations or no response to therapy. Studies reviewing large clinical trials3 8 highlighted predictors of response to anti-IL-5 treatment. In these clinical trials, higher blood eosinophilia seemed to be a better predictor. However, their definition of response was primarily determined by reduction of exacerbation rate and patients were included based on their blood eosinophil counts. Defining eosinophilic asthma using blood eosinophil concentration could lead to misdiagnosis of patients exhibiting sputum eosinophilia (≥3%) without an increase in blood eosinophil numbers, which represents 25% of asthmatic patients.1 In fact, Mukherjee et al. 9 have shown sub-optimal responses when sputum eosinophilia persists despite normalisation of blood eosinophil numbers with mepolizumab treatment. Here, we aimed at evaluating the role of baseline sputum eosinophils in identifying super-responders to mepolizumab and benralizumab. We conducted a prospective study on 106 patients with severe eosinophilic asthma in whom mepolizumab or benralizumab was started at physician’s discretion. Patients were followed in the University Asthma Clinic of Liege between October 2012 and September 2022. Patients fulfilled the reimbursement criteria for mepolizumab and benralizumab in Belgium that require a blood eosinophil count≥300 cells/µL on at least two occasions and a minimum of two exacerbations within the last 12 months. Patients underwent a complete evaluation at baseline and after 24 weeks of treatment. Clinical, functional and inflammatory data were collected at both visits, including sputum cell counts, exacerbation rate, spirometry, chronic OCS use and Asthma Control Questionnaire-7 (ACQ-7). Exacerbations were defined by a course of OCS for at least 3 days for a case of asthma worsening. The study was approved by the Ethics Committee of CHU Liege (Ref. 2009/161), and each patient signed an informed consent. Patients were classified as “super-responders”, “partial responders” or “nonresponders” after 24 weeks of treatment. Super-responders met the following cumulative criteria : no exacerbation in the past 24 weeks; a decrease in ACQ-7 score by at least 0.5 or an ACQ-7 score≤1.5; a diminution of sputum eosinophil count (≥ 50%) or a sputum eosinophil count<3%; an improvement of FEV₁ (post-bronchodilation)≥100 mL or FEV₁ (post-bronchodilation)≥80% and a stop of chronic OCS use after 24 weeks of treatment for patients who required daily maintenance with OCS. Nonresponders were patients who didn’t achieve a decrease of 50% in OCS dose and didn’t reduce their rate of exacerbation (≥ 50%) with mepolizumab or benralizumab therapy after 24 weeks. Finally, partial responders were defined as patients who did not fulfil the criteria of nonresponders and super-responders after 24 weeks of treatment. Comparisons were performed using a Fisher’s exact test for categorical variables, an ANOVA test for continuous variables or a Kruskal-Wallis test when parametric assumptions were not met (table 1). Receiver-operating characteristic (ROC) curve was constructed to assess the ability of the sputum eosinophil count (%) to identify super-responders (figure 1). In this cohort, optimal cut-off point was determined by the Youden index method. A p-value<0.05 was considered statistically significant.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Roc curves representing the ability of sputum eosinophils (%) to identify super-responders to mepolizumab and benralizumab in combined and separate cohorts associated with cut-off points for sputum eosinophil count (%) for predicting a super-response after logistic regression analysis.

The functional, demographic, treatment and inflammatory characteristics at baseline of the severe eosinophilic asthmatic cohort, classified according to their response to benralizumab (n=34) and mepolizumab (n=72) after 24 weeks of treatment are given in table 1. Super-responders represented 30% of the severe eosinophilic asthmatic cohort while partial responders and nonresponders accounted for 60% and 10%, respectively. Super-responders displayed greater baseline sputum eosinophil counts (%) than partial responders and nonresponders (p=<0.01; table 1). The best cut-off point to discriminate super-responders in the whole cohort was 38% of sputum eosinophils (sensitivity, 56%, specificity, 87%, ROC AUC, 0.71, 95% CI, 0.6 to 0.83, p<0.001; figure 1). Specifically, the best cut-off points to discriminate super-responders were 40% of sputum eosinophils for the mepolizumab cohort (sensitivity, 57%, specificity, 86%, ROC AUC, 0.72, 95% CI, 0.59 to 0.86, p=0.003) and 38% of sputum eosinophils (sensitivity, 55%, specificity, 91%, ROC AUC, 0.71, 95% CI, 0.52 to 0.91, p=0.05) for the benralizumab cohort. By contrast, partial responders and nonresponders exhibited a more intense airway neutrophilic inflammation (%) (p<0.01; table 1). Baseline blood eosinophil counts were comparable in all groups, both in absolute value and percentage. Nonresponders were characterised by a greater proportion of bronchiectasis, an absence of nasal polyposis, a more intense smoking history and lower FEV₁ (pre and post,%) values. No difference in baseline FE_NO levels was seen between super-responders and nonresponders. Furthermore, change in FE_NO over time was not significant and similar between the three groups (data not shown).

Our results highlight that airway type-2 inflammation, defined by high sputum eosinophil counts is better than systemic eosinophilia to identify super-responders to mepolizumab and benralizumab. Other real-life studies have examined super-response to mepolizumab and benralizumab6 7 and did not identify blood eosinophils as predictors of super-response. Super-response, as per our definition, is a composite outcome that includes but goes beyond the sole exacerbation component. Our data suggest that when taking into account day to day asthma control and lung function improvement, sputum eosinophil count appears to be a better predictor of response to anti-IL-5(R) than blood eosinophils. Importantly, in our study, sputum eosinophils fared much better than FE_NO to predict a super-response and the latter does not show any significant change after starting anti-IL-5(R), indicating that the molecular pathway leading to FE_NO is clearly independent of IL-5 and eosinophils. In conclusion, our results reinforce the importance to look at airway type-2 inflammatory biomarkers in patients suffering from severe asthma before starting a biologic as it is associated with the intensity of response to mepolizumab and benralizumab.

Main findings of this study

Our study reinforces the importance to examine airway type-2 inflammatory biomarkers in patients suffering from severe asthma before starting a biologic as it is associated with the intensity of response to mepolizumab and benralizumab.