Graphical abstract

In this large, real-world programme, patients with severe eosinophilic asthma treated with benralizumab had substantial improvements in clinical outcomes, irrespective of previous biologic use and key clinical characteristics important to decision-making in clinical practice. #: 3 units Asthma Control Test score or −0.5 units in 6-item Asthma Control Questionnaire score. AER: annualised exacerbation rate; MCID: minimal clinically important difference; mOCS: maintenance oral corticosteroid; Q8W: every 8 weeks.

Abstract

Background Pivotal phase 3 trials and real-world studies have demonstrated benralizumab's overall efficacy and safety in severe eosinophilic asthma (SEA). Additional large-cohort data are needed to confirm its real-world effectiveness in SEA according to previous biologic use and key baseline characteristics important for treatment selection.

Methods XALOC-1 is a large, multinational, retrospective, observational, real-world study programme of benralizumab in adults with SEA. This 48-week integrated analysis assessed annualised exacerbation rate (AER), maintenance oral corticosteroid (mOCS) use, asthma symptom control and lung function during a 12-month baseline period and up to 48 weeks after benralizumab initiation. Subgroup analyses were based on previous biologic use and key baseline clinical characteristics (mOCS use, blood eosinophil count, exacerbation history, age at asthma diagnosis, fractional exhaled nitric oxide level and presence of atopy and chronic rhinosinusitis with nasal polyps).

Results Out of 1002 patients analysed, 380 were biologic-experienced. At week 48, 71.3% were exacerbation-free (versus 17.2% at baseline); relative reduction in AER was 82.7% overall and 72.9% in biologic-experienced patients; rates were maintained across all key clinical characteristic subgroups. Of patients using mOCS at baseline (n=274), 47.4% (130 out of 274) eliminated their use by week 48; the mean reduction from baseline in daily dose was 51.2% and, notably, 34.9% in biologic-experienced patients (n=115). Clinically significant improvements in asthma symptom control and lung function were observed.

Conclusion In this large, real-world programme, SEA patients treated with benralizumab had substantial improvements in clinical outcomes irrespective of previous biologic use and key clinical characteristics important to therapeutic decision-making in clinical practice.

Shareable abstract

Patients with severe eosinophilic asthma who received benralizumab in the real world had substantially improved clinical outcomes, irrespective of prior biologic use and key clinical characteristics, including atopic status and FENO levels https://bit.ly/3TTGveS

Introduction

Approximately 3–5% of adults with asthma have severe refractory disease, characterised by poor symptom control and frequent exacerbations despite adherence to high-dose inhaled therapies, or have adequate asthma control only when taking daily oral corticosteroids (OCS) [1]. >90% of patients with severe asthma are likely to have an eosinophilic phenotype, characterised by eosinophilic airway inflammation and blood eosinophilia [2]. The introduction of biologic treatments that target key mediators of eosinophilic inflammation has therefore provided an opportunity to improve symptom control while also reducing exacerbation risk and OCS dependence in patients with severe refractory asthma [3].

Benralizumab is a humanised monoclonal antibody that binds to the interleukin (IL)-5 receptor (IL-5R) expressed on eosinophils, eosinophilic precursors and basophils, resulting in the rapid, near-complete depletion of these cells through enhanced antibody-dependent cell-mediated cytotoxicity (ADCC) [4, 5]. Based on three pivotal phase 3 randomised controlled trials (RCTs) (SIROCCO, CALIMA and ZONDA), in which benralizumab significantly reduced exacerbation rates, improved lung function and asthma symptoms and reduced daily OCS use versus placebo [6–8], benralizumab was approved as an add-on maintenance treatment for patients with uncontrolled severe eosinophilic asthma (SEA). This benefit was shown to persist for up to 5 years with long-term use [9].

A number of biologic treatments, with varied mechanisms of action, are available for the treatment of patients with SEA [10]. However, claims data from the United States of America found that ∼60% of patients with severe asthma prescribed a biologic (n=3262) had uncontrolled or suboptimally controlled symptoms, highlighting an unmet potential benefit relating to switching biologic treatments [11]. This is compounded by the paucity of data on switching in the real world and the significant underrepresentation of biologic-experienced patients with SEA in clinical trials; in the SIROCCO and CALIMA RCTs, they constituted only 8% and 3% of the study populations, respectively [6, 8]. This represents a considerable limitation when clinicians are discussing treatment options with patients who remain uncontrolled on a biologic. Moreover, the small numbers of patients evaluated in studies, to date, means that effectiveness according to key patient characteristics that might guide treatment decisions has not been evaluated.

The large multinational XALOC programme was established to address these data gaps by assessing the real-world effectiveness of benralizumab, the patterns of its use, and the characteristics of benralizumab-treated patients, both biologic-experienced and biologic-naïve, using retrospective (XALOC-1, five national studies: ANANKE (Italy), BETREAT (Portugal), Benralizumab Patient Access Programme (BPAP) (United Kingdom (UK)), ORBE II (Spain) and VOLTS (Canada)) and prospective (XALOC-2, four national studies) observational data. Initial results of the UK BPAP study and the Italian ANANKE study have already been published [12, 13]; we now report the results of the much larger, 48-week integrated analysis of XALOC-1 from >1000 adult patients with SEA across Canada, Italy, Portugal, Spain and the UK.

As the largest real-world programme to date of patients with SEA treated with benralizumab, including a large proportion of biologic-experienced patients switching to benralizumab, the XALOC programme enables us to describe the real-world effectiveness of benralizumab in greater clarity. This includes an analysis of key baseline characteristics relating to response, providing valuable data for clinicians considering treatment options for patients with uncontrolled SEA despite biologic treatment.

MethodsStudy design

XALOC-1 is a multinational, retrospective, observational, real-world programme composed of five national studies investigating the use of benralizumab in patients with SEA for up to 96 weeks in Canada (VOLTS; n=306), Italy (ANANKE; n=217), Portugal (BETREAT; n=74), Spain (ORBE II; n=204) and the UK (BPAP; n=208) (2018–2023). Eligible patients had received benralizumab according to the approved label and routine clinical practice (supplementary table S1) [14]. The decision to initiate benralizumab was taken independently by the investigator and was not driven by the study protocol. This analysis integrates data from the five individual studies.

The index date was defined as the date of benralizumab initiation; the baseline period was the 12-month period before the index date (supplementary figure S1). The follow-up period for the current analysis was from the index date up to 48±4 weeks (1 year, hereafter referred to as week 48) (supplementary figure S1). Data up to 96 weeks are being analysed separately and will be reported elsewhere. Data were extracted from paper and/or electronic medical records by hospital-/clinic-based healthcare professionals or clinical research organisation employees. Study design, inclusion criteria and data collection (using an electronic case report form platform) were standardised across the five national studies, enabling integration of patient-level data from all studies in a single 48-week analysis (supplementary table S1). Patients might have been lost to follow-up, discontinued the treatment with benralizumab, and/or switched to other biologic(s) after benralizumab initiation. For this 48-week integrated analysis, follow-up of data collection of clinical outcomes was censored at week 48, loss to follow-up, death or switching to another biologic, whichever was earlier. The study did not measure the effect of other biologic(s) other than benralizumab.

All five national studies were performed in accordance with the Declaration of Helsinki, good pharmacoepidemiology practice guidelines, the International Conference on Harmonization guideline for good clinical practice and all applicable legislation on noninterventional studies and/or observational studies. Ethical approval for the studies was provided by the appropriate institutional review board and/or independent ethics committee for each clinical site. The requirement for informed consent varied across the national studies, per local legislation.

Patient population

The inclusion criteria for this integrated analysis were adults (≥18 years) with SEA who received benralizumab as defined in each national study (supplementary table S1) and who had available follow-up data of ≥3 months from the index date. Patients were excluded if they were receiving benralizumab or any other biologic for the treatment of asthma in a clinical trial at the time of enrolment; previous treatment with all other biologic therapies for asthma was allowed.

Patients were categorised as biologic-naïve (no previous biologic treatment for severe asthma recorded during the baseline period), or biologic-experienced (patients who received one or more biologic treatment for asthma during the 12-month baseline period); biologic-experienced patients were further stratified by previous omalizumab or mepolizumab experience (no stratification by previous reslizumab use was performed due to the small sample size; no patients with a history of dupilumab were included, as this treatment was not available at the time of XALOC programme initiation). Additional patient subgroups were defined based on key baseline clinical characteristics: maintenance OCS (mOCS) use, peak blood eosinophil count (BEC), exacerbation history, fractional exhaled nitric oxide (FENO) level, age at asthma diagnosis, atopic status and presence of chronic rhinosinusitis with nasal polyps (CRSwNP) based on medical record data available at the index date (supplementary table S2).

Outcome measures

Outcomes measured at baseline and at weeks 16, 24 and 48 were annualised exacerbation rate (AER), number of asthma exacerbations, mOCS use, mOCS daily dose, asthma symptom control (Asthma Control Test [15] (ACT; Canada, Italy, Portugal, Spain) and 6-item Asthma Control Questionnaire [16] (ACQ-6; Canada, UK)) scores, pre- and post-bronchodilator (BD) forced expiratory volume in 1 s (FEV1), BEC, maintenance asthma treatments other than mOCS and inhaled corticosteroid dose (supplementary table S2). Definitions of asthma exacerbations, controlled asthma and minimal clinically important difference (MCID) in asthma symptom control outcome scores are provided in supplementary table S2. Reasons for discontinuation of previous biologic treatment were captured for the 12-month baseline period, and benralizumab treatment patterns (duration, adherence, discontinuation and reasons for discontinuation) were captured over the 48-week follow-up period. Reasons for discontinuation were based on retrospective data collection from medical records and classified as lack of efficacy (no specific definition beyond less than expected for the biologic after treatment initiation), adverse event (detail not available) or other.

Statistical analysis

Descriptive statistics including mean, median, standard deviation, range for continuous variables, and number and percentage for categorical variables are presented for baseline demographic and clinical characteristics, overall and by patients’ previous biologic use; only nonmissing values are presented. All post-index outcomes are reported using descriptive summary statistics calculated overall, by patients’ previous biologic use and by other key baseline clinical characteristics (lung function outcomes were not reported due to the small sample size) (supplementary table S2).

Longitudinal mixed models with repeated measures were used to summarise continuous end-points over time with estimated means and corresponding 95% confidence intervals for change from baseline in peak BEC, asthma control scores and pre- and post-BD FEV1. The percentage of patients achieving MCID in asthma symptom control outcomes was calculated using binomial Clopper–Pearson exact 95% CIs. AERs (supplementary table S2) and corresponding 95% CIs were calculated for the 12-month baseline period and the follow-up period using generalised linear regression with a negative binomial distribution among patients treated with benralizumab who either discontinued before week 48 or completed 48±4 weeks of follow-up from the index date. A responder analysis was performed on patients who had two or more asthma exacerbations, received mOCS, had an ACQ-6 score >0.75 or an ACT score <20, or had a pre-BD FEV1 of <80% predicted during the 12-month baseline period, and had data for at least one response component available during the 48-week follow-up period. Response was defined as at least one of the following: ≥50% reduction in asthma exacerbations, ≥50% reduction in mOCS dose, change from baseline in the ACQ-6 score ≥–0.5 or ACT score ≥3, or ≥100 mL change in post-BD FEV1 from baseline during the 48-week follow-up. Univariable and multivariable logistic regression analyses were conducted to assess the association between key baseline demographics, clinical characteristics, and responder status at week 48 (supplementary methods).

ResultsPatients

In total, 1002 patients were included in the integrated analysis (Canada: 30.4%, Italy: 21.6%, Portugal: 7.4%, Spain: 19.9%, UK: 20.8%). The mean±sd age at asthma diagnosis was 38.2±18.2 years and 58.6% of patients were female (table 1).

TABLE 1

Baseline demographics, overall and according to patients’ previous biologic experience

Biologic use in the 12-month baseline period

Overall, 37.9% (380 out of 1002) had received at least one biologic treatment in the baseline period: omalizumab, 45.0% (171 out of 380); mepolizumab, 62.4% (237 out of 380); and reslizumab, 8.2% (31 out of 380). The most common reason for discontinuation of previous biologic treatment recorded was lack of efficacy (omalizumab, 57.3% (98 out of 171); mepolizumab, 68.8% (163 out of 237); reslizumab, 58.1% (18 out of 31)). Other reasons for discontinuation included side-effects/adverse events (omalizumab, 5.3% (nine out of 171); mepolizumab, 8.9% (21 out of 237); reslizumab, 16.1% (five out of 31)) and convenience/patient preference (omalizumab, 4.1% (seven out of 171); mepolizumab, 3.8% (nine out of 237); reslizumab, 3.2% (one out of 31)). The median (interquartile range (IQR)) time period between stopping a previous biologic treatment and initiating benralizumab was 45 (1–348) days for omalizumab (n=156), 55 (16–144) days for mepolizumab (n=232) and 56 (43–145) days for reslizumab (n=31).

Baseline disease-related variables were broadly similar between biologic-naïve and biologic-experienced patients, although the IQRs for median BEC and total serum immunoglobulin (Ig)E level were large (table 2).

TABLE 2

Baseline disease characteristics, overall and according to patients’ previous biologic experience

Asthma exacerbations

The relative reduction in AER from baseline to week 48 was 82.7% overall, 87.7% in biologic-naïve patients and 72.9% in biologic-experienced patients (omalizumab-experienced, 75.0%; mepolizumab-experienced, 69.0%) (figure 1, table 3). Overall, 71.3% of patients were exacerbation-free during the follow-up period including 74.9% biologic-naïve patients and 65.4% biologic-experienced patients (omalizumab-experienced, 68.1%; mepolizumab-experienced, 60.4%) (table 3). In comparison, 17.2% of patients overall were exacerbation-free during the 12-month baseline period, including 22.4% of biologic-experienced patients.

FIGURE 1

Relative reduction in annualised exacerbation rate (AER) from baseline to week 48, overall and according to patients’ previous biologic experience and key baseline clinical characteristics. Analyses are based on patients treated with benralizumab who either discontinued before week 48 or completed 48±4 weeks of follow-up from the index date. The number of patients in each subgroup for the analyses is shown. Baseline data relate to exacerbations during the 12-month baseline period prior to the index date. Week 48 data relate to exacerbations during the follow-up period from the index date to week 48±4. mOCS: maintenance oral corticosteroid; BEC: blood eosinophil count; FENO: fractional exhaled nitric oxide; CRSwNP: chronic rhinosinusitis with nasal polyps. #: the most recent measurement in the 12-month baseline period; ¶: based on medical records available at the index date.

TABLE 3

Number of asthma exacerbations and annualised exacerbation rate (AER) at baseline and week 48, overall and according to patients’ previous biologic experience

Relative reductions in AER from baseline to week 48 tended to be higher in patients with higher peak baseline BEC (67.2%, 86.1%, 82.9% and 87.1% for <300, ≥300, ≥300–<500 and ≥500 cells·μL−1, respectively) and higher baseline FENO (78.9%, 83.2% and 85.8% for <20, 20–<50 and ≥50 ppb, respectively) (supplementary table S3). Relative reductions in AER were 84.9% and 81.9% in patients with (n=264) and without concomitant CRSwNP (n=584); 83.0% and 90.8% in patients with positive (n=380) and negative atopic status (n=158); and 80.1% and 85.6% in patients with (n=466) and without (n=381) mOCS use, respectively, during the 12-month baseline period.

Maintenance OCS use

Of the 322 (32.1%) patients receiving mOCS at the index date, 274 patients had week 48 follow-up data available for mOCS. Among these patients, marked reductions in the median daily mOCS dose were observed (figure 2a) from 10.0 mg·day−1 at the index date to 2.5 mg·day−1 at week 48. Additionally, 47.4% eliminated mOCS use completely (daily dose reduced to 0 mg·day−1) by week 48, including 56.1% and 36.5% of biologic-naïve and biologic-experienced patients, respectively (figure 2b, supplementary figure S2). The mean percentage change in daily mOCS dose from index date to week 48 among patients using mOCS at the index date was −51.2%, −63.2% and −34.9%, for patients overall, biologic-naïve patients and biologic-experienced patients, respectively (figure 3, supplementary table S4). Across all other subgroups of the overall population defined by key baseline clinical characteristics, the mean percentage change in daily mOCS dose was between −31.3% and −69.9%.

FIGURE 2

a) Median maintenance oral corticosteroid (mOCS) daily dose at baseline and week 48, and b) percentage of patients with a 100% reduction in mOCS dose from baseline to week 48, overall and according to patients’ previous biologic experience, among patients who were using mOCS at the index date. Data are based on patients who were using mOCS at the index date and had follow-up data at week 48. The box and whisker plots show the medians, interquartile ranges and minimum and maximum values for mOCS daily dose. #: mOCS dose was calculated as the patient's mean daily dosage over the past 30 days on or prior to the target date of the specified visit (the index date or week 48).

FIGURE 3

Mean percentage change in maintenance oral corticosteroid (mOCS) dose from baseline to week 48, overall and according to patients’ previous biologic experience and key baseline clinical characteristics, in patients using mOCS at the index date. mOCS dosage was calculated as a patient's mean daily dosage over the past 30 days on or prior to the target date of the specified visit (the index date or week 48). The number of patients with mOCS use at the index date and with follow-up data at week 48 is shown. BEC: blood eosinophil count; FENO: fractional exhaled nitric oxide; CRSwNP: chronic rhinosinusitis with nasal polyposis. #: most recent measurement in the 12-month baseline period. ¶: based on medical records available at the index date.

Asthma symptom control scores

For patients overall, changes from baseline to week 48 in least-squares mean (95% CI) ACQ-6 and ACT scores were −0.82 (–1.00– −0.64; n=226) and 6.48 (5.81–7.16; n=294), respectively. At week 48, 67.8% patients had improvements matching or exceeding the MCID of 3 units for the ACT score or of −0.5 units for the ACQ-6 score (72.0% and 60.9% in biologic-naïve and biologic-experienced subgroups, respectively) (figure 4). Most patients in all other subgroups defined by key baseline clinical characteristics achieved the MCID for ACT or ACQ-6 score at week 48 (figure 4, supplementary table S5). Additional outcomes for asthma symptom control at weeks 16, 24 and 48 are presented in supplementary figure S3.

FIGURE 4

Percentage of patients with improvements matching or exceeding the minimal clinically important difference in Asthma Control Test (ACT) score (3 units) or 6-item Asthma Control Questionnaire (ACQ-6) score (–0.5 units) from baseline to week 48, overall and according to patients’ previous biologic experience and key baseline clinical characteristics. Baseline ACQ-6 and ACT scores were measured at the index date or were the most recent measurement in the 12-month baseline period prior to the index date. The number of patients with ACQ-6 or ACT score data at baseline and week 48±4 is shown. Analyses are based on patients with ACQ-6 or ACT score data at baseline and week 48±4. mOCS: maintenance oral corticosteroid; BEC: blood eosinophil count; FENO: fractional exhaled nitric oxide; CRSwNP: chronic rhinosinusitis with nasal polyps. #: at the index date or the most recent measurement in the 12-month baseline period; ¶: based on medical records available at the index date.

Lung function

Overall, 263 out of 1002 patients had data for baseline and at least one post-baseline pre-BD FEV1 measurement. At week 48, the least-squares mean change (95% CI) from baseline in pre-BD FEV1 was 0.27 (0.20–0.34) L overall (n=263), 0.27 (0.18–0.36) L in biologic-naïve patients (n=176) and 0.29 (0.18–0.39) L in biologic-experienced patients (n=87). Improvements in pre-BD FEV1 from baseline to week 48 of ≥100 mL and ≥200 mL were seen in 58.4% (66 out of 113) and 48.7% (55 out of 113) of patients overall, respectively, with similar improvements observed among biologic-naïve and biologic-experienced patients (supplementary figure S4). Improvements in post-BD FEV1 from baseline to week 48 of ≥100 mL and ≥200 mL were seen in 51.9% (96 out of 185) and 40.0% (74 out of 185) of patients, respectively; improvements among biologic-naïve and biologic-experienced patients were similar.

Persistence of treatment

Overall, 77.9% of patients were still receiving benralizumab at week 48 (77.0% in the biologic-naïve and 79.5% in the biologic-experienced subgroups) (supplementary figure S5a). The median (IQR) treatment durations were 352 (184–624) days, 366 (225–629) days and 344 (169–609) days for patients overall, biologic-naïve and biologic-experienced patients, respectively.

Among patients no longer receiving benralizumab at week 48 (n=221), 6.8% discontinued at <3 months, 27.6% at 3–6 months and 65.6% at 7–12 months. More patients discontinued benralizumab at <3 months in the biologic-experienced than biologic-naïve group (9.0% versus 5.6%; supplementary figure S5b). The most commonly reported reasons for discontinuation after 6- and 12-months of follow-up were lack of efficacy (45.2% (14 out of 31); 43.5% (54 out of 124)) or adverse events (25.8% (8 out of 31); 23.4% (29 out of 124)), respectively.

There was no change in the percentage of patients receiving maintenance asthma treatment, other than mOCS, between baseline and week 48.

There was near-complete depletion of peak BEC between baseline and week 48, regardless of previous biologic use (supplementary table S6).

Responder analysis

A total of 690 patients were evaluable for response, most of whom (612 out of 690 (88.7%)) were considered responders at week 48. Univariable logistic regression analysis on the relationship between pre-specified baseline demographics, key clinical characteristics and responder status at week 48 highlighted that at baseline, responders were more likely to have a greater number of exacerbations, higher peak BEC, lower mOCS daily dose, prior biologic experience, CRSwNP and have negative atopic status compared with nonresponders (supplementary table S7).

Multivariable logistic regression analyses evaluating the relationship between pre-specified baseline demographics, key clinical characteristics, and responder status at week 48 highlighted that a greater number of exacerbations (OR 1.54, 95% CI 1.15–2.07; p=0.004) and higher peak BEC (OR 1.48, 95% CI 1.05–2.08; p=0.025) at baseline were positively associated with response at week 48 (supplementary table S7).

Discussion

XALOC-1 is the largest real-world programme of benralizumab to date, offering several novel insights both in relation to biologic-experienced and biologic-naïve patients with SEA initiating benralizumab. There are currently very limited data on biologic-experienced patients due to poor representation in RCTs and currently published smaller real-world analyses. The scope and size of the XALOC-1 programme, which includes patients from five countries, enabled comprehensive and meaningful evaluation of the real-world effectiveness of benralizumab in almost 400 biologic-experienced patients, most of whom had discontinued their previous biologic therapy because of lack of efficacy. Approximately two-thirds of biologic-experienced patients were exacerbation-free at 1 year; one-third had eliminated mOCS use; and >60% had a clinically meaningful improvement in asthma symptom control. Many biologic-experienced patients who were dependent on mOCS at initiation were able to reduce their daily mOCS dose substantially, with the mean daily mOCS dose reduced by approximately one-third. Benralizumab was effective, irrespective of whether patients switched from mepolizumab or omalizumab. Mechanistically, the data suggest that inadequate eosinophil suppression with these biologic therapies is a clinically relevant issue in a subgroup of patients with SEA and provides support for the use of benralizumab in a broad group of patients suboptimally controlled on alternative biologic therapies.

In the overall population of >1000 adult patients, benralizumab was effective in reducing AER and mOCS use, and improving asthma control and lung function. After 1 year of benralizumab treatment, >70% of patients were exacerbation free, nearly one-half had eliminated mOCS use and two-thirds showed clinically meaningful improvements in asthma control based on ACT or ACQ-6 scores. The magnitude of these outcomes across a large multinational patient population emphasises the real-world value of benralizumab. While biologic-experienced patients tended to have less pronounced improvements than biologic-naïve patients, the effectiveness was notable and clinically relevant. These findings are supported by other smaller real-world studies of benralizumab in severe asthma [12, 17, 18] and should increase physician confidence when considering whether to switch patients with SEA to benralizumab from another biologic treatment.

Our results show the benefits of switching biologic treatments in patients with SEA that remains uncontrolled in the real world; specifically, that patients who experience a suboptimal response to an anti-IL-5 treatment can still benefit from an IL-5R-targeted treatment, suggesting the residual eosinophilic inflammation on anti-IL-5 is clinically relevant in some patients. Benralizumab, mepolizumab and reslizumab all inhibit IL-5 signalling, which is key to the maturation, activation and survival of eosinophils [17]. However, benralizumab triggers eosinophil apoptosis through ADCC due to its afucosylated Fc region [4, 18], which results in the rapid and complete or near-complete depletion of blood eosinophils [5], as well as significant depletion of eosinophils and eosinophil lineage-committed progenitor cells in the airways and sputum of patients with asthma [19, 20]. Treatment with the anti-IL-5 monoclonal antibody mepolizumab results in a more modest reduction in BEC [5], and fewer patients achieve complete elimination [21]. This is because there are other non-IL-5 essential drivers of eosinophil biology, including IL-3 and granulocyte–macrophage colony-stimulating factor, which may explain the persistence of residual airway eosinophils and ongoing degranulation reported with mepolizumab treatment [22, 23], and the persistence of sputum eosinophilia in nearly 50% of exacerbations in one study of mepolizumab-treated patients [24].

Improvements in clinical outcomes among omalizumab-experienced patients who received benralizumab in the XALOC-1 programme support other evidence suggesting that the presence of atopy and sensitisation to aeroallergens does not necessarily mean that IgE is the optimal therapeutic target. This was first shown in patients switching from omalizumab to mepolizumab [25]. Subsequent data indicate that the efficacy of benralizumab is independent of total IgE level [26], and that patients eligible for omalizumab treatment respond similarly well to an anti-IL-5/-5R monoclonal antibody as patients not eligible for omalizumab [27].

The benefits of benralizumab in XALOC-1 were maintained in all patient subgroups defined by key baseline clinical characteristics used to guide clinical treatment selection, notwithstanding some differences in the magnitude of response. This reinforces the broad therapeutic applicability of benralizumab. Consistent with RCT post hoc and exploratory analyses [28, 29], and the mode of action of benralizumab, response to treatment was more pronounced in patients with higher BEC. However, AER reductions of approximately >70% were still observed in all baseline BEC subgroups. This could be explained by the ongoing anti-eosinophilic effect of either anti-IL-5 treatment in patients who previously received mepolizumab or reslizumab, or from the eosinophil-lowering impact of mOCS treatment. Indeed, this further highlights the importance of contextualising BEC measurements as was proposed in a recent expert consensus framework [2]. Failure to do this risks incorrectly labelling patients as “noneosinophilic” and preventing potentially transformational treatments from being initiated.

Benralizumab was effective in reducing AER regardless of baseline FENO, and the response improved numerically by small increments with increasing baseline FENO level. FENO reflects IL-13 activity, and IL-13 induces the key eosinophil chemoattractant CCL26 [30]. Thus, FENO in part reflects the chemoattractive force drawing eosinophils into the airways, and so it is not unexpected that patients with the highest FENO appeared to have the greatest numerical improvements with benralizumab treatment. Additionally, eosinophils are a source of IL-13, and it is possible that some of the efficacy of benralizumab may be due to a direct or indirect reduction in airway IL-13 following eosinophil depletion [31, 32].

Many of the clinical characteristics we observed to be associated with response to benralizumab are consistent with those seen in post hoc analyses of the phase 3 benralizumab trials as well as other smaller real-world cohorts, and reflect features seen in more eosinophilic patients [28, 33]. These include a greater number of exacerbations and higher BEC at baseline.

At the most severe end of the asthma spectrum remains a significant proportion of patients who unfortunately continue to rely on a daily dose of mOCS to avoid a deterioration in asthma control [34, 35]. However, chronic OCS exposure has a very poor safety profile and is associated with numerous and adverse outcomes including osteoporosis, type II diabetes, pneumonia and cataracts [36, 37]. In the prospective PONENTE and ZONDA clinical trials of OCS-dependent patients with SEA, 63% and 52% of patients treated with benralizumab, respectively, were able to eliminate mOCS [7, 38]. In the real-world setting of XALOC-1, patients with SEA could on average halve their mean daily mOCS dose, and ∼50% could stop their mOCS altogether. Further reduction of mOCS use may not have been possible due to adrenal insufficiency rather than persistent OCS-responsive type 2 inflammation.

The size of the patient cohort, the inclusion of both biologic-experienced and biologic-naïve patients and the amount of clinical information collected have allowed us to describe outcomes within patient subgroups in a way that was not possible in other real-world analyses to date. National prospective and retrospective studies have demonstrated similar real-world benefits of benralizumab treatment, but may have been limited by small patient populations, few biologic-experienced patients, shorter follow-up duration, or restriction to a single centre [12, 13, 33, 39–43]. The integrated XALOC-1 analysis gives greater confidence in the generalisability of these benralizumab data, including in biologic-experienced patients. The real-world setting of XALOC-1 was important to ensure a more diverse patient population than can be represented in clinical trials. The main limitations of this study are due to its retrospective design, the lack of a control arm, and being restricted to data available in routine clinical practice. For example, there were fewer data for patient-reported outcomes and lung function, and the latter may have been biased towards patients with more severe disease, as they may be more likely to have lung function data measured at regular intervals. In addition, there is a possibility that some biologic-experienced patients may have been incorrectly categorised as biologic-naïve, owing to previous biologic use status only being assessed during the pre-defined 12-month baseline period.

In summary, in a real-world international cohort of >1000 patients with SEA, including 380 biologic-experienced patients who switched biologic treatment, XALOC-1 demonstrates that benralizumab treatment is highly effective, including in patients who have not adequately responded to anti-IgE and/or anti-IL-5 treatments, irrespective of key clinical characteristics such as BEC and FENO level.

Supplementary materialSupplementary Material

Please note: supplementary material is not edited by the Editorial Office, and is uploaded as it has been supplied by the author.

Supplementary material ERJ-01521-2023.Supplement

Acknowledgements

This study was supported by AstraZeneca. The authors would like to thank Susanna Ryan and Anna Mett (inScience Communications, Springer Healthcare, London, UK), for providing medical writing support, which was in accordance with Good Publication Practice (GPP 2022) guidelines (https://www.acpjournals.org/doi/10.7326/M22-1460) and funded by AstraZeneca. The authors would like to thank Tina Wu, Sisi Gu, Yifei Gao and Dipali Shah (ClinChoice Inc., Fort Washington, PA, USA), who conducted the data integration of the XALOC-1 programme studies and data analysis. We would like to thank the following people: Stephen G. Noorduyn (GSK, Cambridge, ON, Canada (formerly of AstraZeneca, Mississauga, ON, Canada)), and Jesús Sanchez Tena (AstraZeneca, Madrid, Spain) for their work on protocol design and local study coordination; Andrew Menzies-Gow (AstraZeneca, Cambridge, UK) and Justin Kwiatek (GSK, Upper Providence, PA, USA (formerly of AstraZeneca, Wilmington, DE, USA)) for their valuable contributions to data interpretation and discussion; Catherine Bragg (AstraZeneca, Cambridge, UK) for her valuable contribution in driving and leading the medical writing support; Michał Rupert (AstraZeneca, Warsaw, Poland) and Surendra Saimpu (AstraZeneca, Cambridge, UK) for their valuable contribution to XALOC data integration; Kanthilatha Chitikireddi (AstraZeneca, Warsaw) for her valuable operational contribution to the XALOC programme; Chris Brooks (AstraZeneca, Cambridge, UK) and Carrie Lancos (AstraZeneca, Gaithersburg, MD, USA) for their previous contributions to medical writing support for XALOC; and Marianna Alacqua (CSL Behring, Milan Italy (formerly of AstraZeneca, Cambridge, UK)) for her previous contribution to the XALOC programme design. We acknowledge with gratitude the support and contributions made by other individuals involved in the XALOC-1 study programme at the country level. A full list of contributors is provided in the supplementary material.

Footnotes

Ethics statement: Ethical approval for the studies was provided by the appropriate institutional review board and/or independent ethics committee for each clinical site.

Data sharing statement: Data underlying the findings described in this manuscript may be obtained in accordance with AstraZeneca's data sharing policy described at https://astrazenecagrouptrials.pharmacm.com/ST/Submission/Disclosure.

This article has an editorial commentary: https://doi.org/10.1183/13993003.00716-2024

Conflict of interest: D.J. Jackson has received consultancy fees and speakers’ fees from AstraZeneca, Boehringer Ingelheim, Chiesi, GlaxoSmithKline and Sanofi Regeneron, and research grants from AstraZeneca. G. Pelaia has received lecture fees and advisory board fees from AstraZeneca, Boehringer Ingelheim, Chiesi, GlaxoSmithKline, Guidotti, Insmed, Lusofarmaco, Menarini, Neopharmed Gentili, Novartis, Sanofi and Zambon. B. Emmanuel, T.N. Tran, D. Cohen, V.H. Shih, A. Shavit, M. Watt, S. Kayaniyil, M. Pardal, D. Arbetter and A.P.J. Rabe are employees of, and own stock in, AstraZeneca. S. Boarino and J. Nuevo are employees of AstraZeneca. R. Katial (currently of National Jewish Health and University of Colorado Denver, Denver, CO, USA) and E. Garcia-Gil (currently of Almirall, Barcelona, Spain) were employees of AstraZeneca at the time the study was conducted. C. Chaves Loureiro has received consultancy fees and speakers’ fees from AstraZeneca, Chiesi, GlaxoSmithKline, Sanofi Regeneron and Teva, and research grants from GlaxoSmithKline. A. Padilla-Galo reports grants, personal fees and non-financial support from AstraZeneca and Sanofi, personal fees, and non-financial support from Chiesi, GlaxoSmithKline, Novartis and Teva, and personal fees from ALK, Bial and FAES, outside the submitted work. P. Nair reports that in the past two years, his institution received grant support from AstraZeneca, Cyclomedica, Equillium, Foresee, Genentech, Sanofi and Teva; he has also received honoraria from Arrowhead, AstraZeneca, CSL Behring, GlaxoSmithKline and Sanofi.

Support statement: This study was supported by AstraZeneca. Funding information for this article has been deposited with the Crossref Funder Registry.

Received September 8, 2023.Accepted March 20, 2024.