A four-year journey towards clinical remission and background therapies optimization with benralizumab: A retrospective analysis

Abstract

Background

Clinical remission is increasingly recognized as a primary treatment goal in severe asthma, particularly severe eosinophilic asthma (SEA), where comorbidities such as chronic rhinosinusitis with nasal polyps (CRSwNP) contribute to persistent inflammation and treatment burden. Benralizumab, an anti–IL-5Rα monoclonal antibody, has shown efficacy in achieving sustained control in SEA, but long-term real-world data beyond 36 months remain limited.

Objective

To evaluate the long-term (48-month) effectiveness of benralizumab in SEA patients, focusing on clinical remission, airway function, and reduction of background therapy, including oral corticosteroids (OCS) and inhaled corticosteroids (ICS).

Methods

In this retrospective, multicenter study across 9 Italian centers, 128 adult SEA patients (mean age 57.1 years, 57% female; 59.4% with CRSwNP) were treated with subcutaneous benralizumab for up to 48 months. Clinical remission was assessed using SANI criteria: partial (pCR, meeting ≥2 of 3 criteria without OCS) and complete (cCR, all criteria fulfilled without OCS). Outcomes included annualized exacerbation rate (AER), pulmonary function (FEV₁, FEV₁/FVC, FEF25-75%), airway inflammation (FeNO, blood eosinophils), patient-reported outcomes (ACT, ACQ-6, AQLQ), and background therapy use. Mixed-effects regression models evaluated longitudinal changes.

Results

Benralizumab induced rapid and sustained improvements. AER decreased from 3.62 to 0.33 at 48 months; blood eosinophils were nearly undetectable throughout. FEV₁ increased from 2.04 L to 2.49 L, with proportional gains in FEV₁/FVC and FEF25-75%. ACT and ACQ-6 scores improved significantly (p < 0.001), and sinonasal symptoms in CRSwNP patients declined (VAS 4.70 → 2.31). OCS discontinuation was achieved in 88.5% of patients at 48 months, with mean ICS dose reduced by 35%. Overall, clinical remission was observed in 87.9% of patients, with cCR in 63.8%. Improvements were consistent regardless of ICS dose reduction.

Conclusions

Long-term benralizumab treatment in SEA patients leads to rapid, sustained clinical remission, improved large and small airway function, and significant reduction of background therapy, including OCS and ICS. These results support the role of benralizumab as a disease-modifying therapy and highlight the feasibility of remission-based management in real-world practice.

Introduction

In severe asthma, clinical remission is gaining recognition as a primary treatment objective, reflecting a transition from symptom management toward sustained disease modification.^1,2 Clinical remission is generally defined by the absence of exacerbations, withdrawal of oral corticosteroids (OCS), stable lung function, and optimal symptom control, thereby providing a pragmatic and patient-centered benchmark of therapeutic success.^3,4 This approach aligns with the growing "treat-to-target" paradigm and the broader recognition of disease-modifying anti-asthmatic drugs (DMAADs) as key tools to achieve sustained control of airway inflammation and reduce treatment burden.^5,6

Severe eosinophilic asthma (SEA) is a phenotype of severe asthma characterized by high levels of eosinophils in the blood or sputum.^7,8 This form of asthma often leads to more frequent exacerbations, declining lung function, and a greater necessity of oral corticosteroids,⁹ resulting in significant morbidity, lower quality of life,¹⁰ and increased healthcare costs.^11,12 The burden of SEA, therefore, remains substantial and can be further worsened by the presence of comorbidities, among which chronic rhinosinusitis with nasal polyposis (CRSwNP) is especially common.^13,14 CRSwNP contributes to upper and lower airway inflammation and is associated with lower asthma control, increased symptoms burden, and a higher frequency of exacerbations.

Despite adhering to high-dose inhaled corticosteroids (ICS) and long-acting β2-agonists (LABA), many patients experience persistent symptoms and rely on systemic corticosteroids, which exposes them to well-documented adverse effects 8, 9, 10. Optimizing background therapy, particularly minimizing or eliminating OCS, remains a key unmet need in SEA management.

To support standardized remission assessment, the Severe Asthma Network Italy (SANI) introduced a definition that includes both complete clinical remission (cCR), defined as the absence of OCS use and fulfillment of 3 criteria (no exacerbations, no symptoms, stable lung function), and partial clinical remission (pCR), which requires 2 of the 3 in the absence of OCS.⁴

This framework has been widely adopted in Italian real-world studies and has facilitated the evaluation of remission outcomes across different biologic therapies and patient populations.15, 16, 17, 18

Benralizumab is a humanised, afucosylated IgG1κ monoclonal antibody directed against the interleukin-5 receptor alpha (IL-5Rα). Through enhanced antibody-dependent cell-mediated cytotoxicity, it induces rapid and near-complete depletion of eosinophils, resulting in sustained suppression of eosinophilic inflammation. 19, 20, 21, 22 Its efficacy and safety have been demonstrated in phase III trials and their long-term extensions, as well as in several real-world studies. 23, 24, 25, 26, 27, 28, 29

Building on our previous 36-month analysis,¹⁵ which demonstrated high rates of clinical remission in patients with SEA treated with benralizumab, the present study evaluates outcomes after up to 48 months of treatment. Given the limited availability of real-world data beyond 3 years, we conducted a retrospective multicenter analysis to assess long-term clinical remission, lung function, and optimization of background asthma therapy, including oral and inhaled corticosteroids.

Methods

Study design

This was an observational, retrospective, multicenter study carried out across 9 specialized centers in Italy affiliated with the Severe Asthma Network in Italy (SANI). The study aimed to evaluate the long-term effectiveness of benralizumab in patients with severe eosinophilic asthma (SEA) over a follow-up period of up to 48 months. Data were retrospectively extracted from medical records between January 2018 and February 2022. The study was conducted in compliance with the Declaration of Helsinki and received approval from the central ethics committee of the SANI network. Written informed consent was obtained from all patients.

Patient population

Adult patients with a diagnosis of SEA were included based on the European Respiratory Society (ERS) and American Thoracic Society (ATS) criteria for severe asthma. Eligibility for benralizumab treatment followed the Italian Medicines Agency (AIFA) therapeutic plan. Patients were required to have a blood eosinophil count (BEC) ≥300 cells/μL in the absence of oral corticosteroids (OCS), and 1 of the following: (1) at least 2 exacerbations in the prior 12 months despite high-dose inhaled therapy (GINA steps 4–5) or (2) continuous OCS therapy in the preceding year.

Benralizumab was administered subcutaneously at a dose of 30 mg every 4 weeks for the first 3 doses, followed by dosing every 8 weeks. Patients were enrolled from centers in Brescia, Catania, Modena, Montebelluna, Padova, Varese, Verona, Siena, and Rome. Follow-up visits were conducted at 6, 12, 24, 36, and 48 months after treatment initiation.

Outcome measures

Baseline and follow-up assessments included sociodemographic data, clinical history, comorbidities, and asthma treatment regimens (including ICS, LABA, OCS, and prior biologics). Laboratory data included BEC, fractional exhaled nitric oxide (FeNO), and total serum IgE. Lung function was evaluated using pre-bronchodilator spirometry (FEV1, FVC, FEV1/FVC, FEF_25–75%). Pulmonary airway obstruction (PAO) was evaluated using a FEV1/FVC threshold of 70%, whereas small-airway dysfunction (SAD) was assessed using a FEF_25–75% predicted threshold of 50%.

Asthma control and quality of life were assessed using the Asthma Control Test (ACT), Asthma Control Questionnaire-6 (ACQ-6), and Asthma Quality of Life Questionnaire (AQLQ). In patients with comorbid chronic rhinosinusitis with nasal polyps (CRSwNP), sinonasal symptoms were evaluated using the Visual Analogue Scale (VAS) and the Sinonasal Outcome Test 22 (SNOT-22). Exacerbations were recorded and analyzed as annualized exacerbation rate (AER).

Clinical remission (CR) was assessed using the criteria established by the SANI Delphi consensus. Partial CR (pCR) was defined as no OCS use and fulfillment of at least 2 of the following: ACT ≥20, no exacerbations, or stable pulmonary function. Complete CR (cCR) required all of the following: no OCS use, ACT ≥20, absence of exacerbations, and pulmonary stability. For our analyses, we defined stable lung function as a consistent improvement in FEV₁ of ≥150 mL from baseline, in line with previous real-world studies on remission with biologics,^15,30 to ensure comparability with existing literature.^4,6

Statistical analyses

For descriptive analyses, continuous variables were given as the mean with standard deviations (SD), or a median with range or interquartile range (IQR), and categorical variables were expressed as the number of subjects (n) and percentage values. Pulmonary airway obstruction (PAO), also referred to as fixed airflow obstruction (FAO), was defined as a post-bronchodilator FEV₁/FVC ratio < 0.70,³¹ in accordance with prior studies in severe asthma populations.³² This threshold, while not specified in GINA guidelines, has been used to characterize persistent airflow limitation in real-world and observational asthma studies. Small-airway dysfunction (SAD) was assessed using a FEF_25–75% predicted cut-off of <50%.³³

Possible overdispersion for count values was assessed using a formal test based on the code from Gelman and Hill (2007).³⁴ Linear, logistic, and Poisson mixed-effects regression models were used to evaluate changes in AER, lung function, PROs scores, OCS, and the use of asthma medication over time, respectively, for continuous, dichotomous, and count values. Regression coefficients, odds ratios (ORs), and exponential regression coefficients associated with each outcome were calculated, along with 95% confidence intervals (CIs), for each factor.

Centre and subject variability were considered as random effects in all mixed-effects regression models. The likelihood ratio test was used as a test of statistical significance, and p-values were adjusted for multiple comparisons using the Holm correction method.

Differences with a p-value less than 0.05 were considered significant. Data were acquired and analyzed in the R v4.3.1 software environment.³⁵

Results

Patients’ characteristics at baseline

Baseline demographic, clinical, and functional characteristics of the studied population are summarized in Table 1. A total of 128 patients were enrolled for the study, with a mean age of 57.1 years and a female representation of 57.0%. Body mass index (BMI) data were available for 126 patients; 46.0% had a normal or underweight BMI, 33.3% were overweight, and 20.7% were classified as obese. Most patients were non-smokers (68.7%), while 29.7% were former smokers and only 1.6% were current smokers.

Table 1.

Demographic, social, and clinical characteristics of patient population at baseline.

Characteristic at Baseline	Patient Population
Age at start benralizumab	57.1 (11.6)
Gender (female)	73 (57.0%)
BMI(n=126)
Under- Normal BMI	58 (46.0%)
Overweight	42 (33.3%)
Obesity	26 (20.7%)
Smoking status
Non-smokers	88 (68.7%)
Previous smokers	38 (29.7%)
Current smokers	2 (1.6%)
Comorbidities
CRS (n=72)	26 (36.1%)
CRSwNP (n = 123)	73 (59.4%)
Allergic rhinitis (n=8)	0 (0%)
Gastroesophageal reflux disease (GERD) (n=123)	45 (36.6%)
Bronchiectasis (n=125)	33 (26.4%)
Eosinophilic esophagitis (n=92)	4 (4.3%)
Age at asthma diagnosis (years; n=104)	35.0 (15.3)
Prior biological treatment (n = 115)
Naïve	87 (75.7%)
Switch	28 (24.3%)
ICS dose (μg/day)(n=110)
<500	23 (20.9%)
500-1000	51 (46.4%)
>1000	36 (32.7%)
LAMA(n=116)	86 (74.1%)
Anti LT(n=124)	53 (42.7%)
Theophylline(n=124)	6 (4.8%)
LABA(n=118)	118 (100%)
OCS(mg/day)	67 (52.3%)
OCS
<=5 mg/day	30 (44.8%)
>5 mg/day	37 (55.2%)
EOS (cells/μL)å (n = 117)	480 [0, 2400]
FeNO (ppb) (n = 62)	55.6 (53.9)
Total serum IgE (IU/mL) (n=92)	322 (399)
AER(n=125)	3.62 (3.2)
Exacerbation phenotype(n=40)
Infectious	29 (72.5%)
Bacterial	17 (58.6%)
Viral	12 (41.4%)
Non Infectious	11 (27.5%)
Lung function
FEV1 (L/min) (n=112)	2.04 (0.9)
FVC (n=110)	3.16 (1.1)
FEV1% pred. (n=110)	75.1 (22.3)
FVC% Pred. (n=111)	91.9 (22.8)
FEV1/FVC (n=111)	57.6 (23.5)
≤70	81 (72.9%)
FEF 25–75% pred (n=82)	46.9 (31.7)
≤50	52 (63.4%)
Asthma PROs
ACT Score (n=116)	15.2 (5.3)
ACQ score (n=47)	6.02 (8.0)
AQLQ score (n=20)	4.48 (1.3)
CRSwNP PROs
VAS (n=49)	4.70 (3.3)
SNOT-22 (n=57)	45.9 (21.9)

Demographic, social, and clinical characteristics of the patient population at baseline. Data refer to n = 128 patients, unless otherwise specified, and are expressed as the number of subjects (percentage), mean (SD), or median (range) as appropriate

Comorbidities were highly prevalent. Chronic rhinosinusitis with nasal polyposis (CRSwNP) was diagnosed in 59.4% of patients, and 36.6% were affected by gastroesophageal reflux disease (GERD). Bronchiectasis was reported in 26.4% of the population, 5.2% had confirmed eosinophilic granulomatosis with polyangiitis (EGPA), and eosinophilic esophagitis was recorded in 4.3%. Only 8 patients had data available on allergic rhinitis, and none of these were positive.

Patients were diagnosed with asthma at a mean age of 35.0 years. Three-quarters of patients (75.7%) were naïve to biological therapy, while 24.3% had previously been treated with other biologics. All patients were being treated with LABA (long-acting beta-adrenoceptor agonists). The majority required additional maintenance therapy, including LAMA (long-acting muscarinic antagonist) (74.1%) and anti-leukotrienes (42.7%), while 4.8% were on theophylline. Concerning the daily dose of inhaled corticosteroids (ICS), 32.7% of patients were taking high-dose ICS (>1000 μg/day), 46.4% on medium-dose (500–1000 μg/day), and 20.9% on low-dose ICS (<500 μg/day). More than half of the patients (52.3%) were on maintenance oral corticosteroids (OCS), with 55.2% of those receiving >5 mg/day.

Baseline values included a median Blood Eosinophil Count (BEC) of 480 cells/mm³ (range 0–2400; n = 117), a mean FeNO of 55.6 ppb (n = 62), and a median total IgE of 322 IU/mL (n = 92). The mean annualized exacerbation rate (AER) was 3.62 (3.2). Among the 40 patients whose exacerbation phenotype was assessed, 72.5% were classified as infectious (58.6% bacterial, 41.4% viral), while 27.5% of exacerbations were non-infectious. Basal lung function was found to be impaired, with a mean FEV1 of 2.04 L and a mean predicted FEV1 of 75.1%. The mean FEV1/FVC ratio was 57.6%, with 72.9% of patients having a ratio below the 70% threshold. Small airway impairment was also relevant, with a mean FEF_25–75% predicted of 46.9%, and 63.4% of patients showed values ≤ 50%.

Patient-reported outcomes (PROs) indicated suboptimal asthma control, with a mean ACT score of 15.2 (n = 116), ACQ of 6.02 (n = 47), and AQLQ of 4.48 (n = 20). In patients with CRSwNP, sinonasal symptom burden was moderate, as reflected by a mean VAS of 4.70 (n = 49) and a mean SNOT-22 score of 45.9 (n = 57).

Asthma exacerbation reduction and inflammatory markers

Among patients with SEA treated with benralizumab, there was a rapid and sustained reduction of exacerbation rates over the 48-month observation period (Fig. 1A). AER at the baseline declined from a mean of 3.62 (3.17) to 0.41 (1.00) (estimate (95% CI) = 0.10 (0.07:0.13)) already after 6 months of treatment and remained low at all time points, reaching 0.33 (0.65) (estimate (95%CI) = 0.09 (0.06:0.13)) at 48 months.

Fig. 1.

Change in AER (A) and BEC (B) during the treatment with benralizumab. Data were recorded at baseline and at 6, 12, 24, 36, and 48 months. Mean (SD), n values, and exponential beta regression coefficients (ie, ratio) with 95% CI, or median (IQR) and n values are reported for each time point.

Consistent with its mechanism of action, benralizumab induced a near-complete depletion of blood eosinophils, as shown in Fig. 1B. Median BEC fell from 480 cells/mm³ (range 0–2400) at baseline to 0 cells/mm³ at 6 months and remained undetectable or near-undetectable throughout the 48-month treatment period. This profound suppression of eosinophils confirms the persistent biological effect of benralizumab on the eosinophilic inflammatory pathway.

Lung function

Patients treated with benralizumab showed a sustained improvement in lung function over the 48-month treatment period (Fig. 2). The mean pre-bronchodilator FEV₁ volume increased from 2.04 L (0.85) at baseline to 2.37 L at 6 months, reaching a peak of 2.44 L at 12 months, and remaining stable thereafter (2.43 L at 24 months, 2.46 L at 36 months, and 2.49 L at 48 months).

Fig. 2.

Change in lung function (FEV₁ volume (A) and percentage of predicted (B), FVC percentage of predicted (C), FEV₁/FVC (D), and FEF_25–75% (predicted) (E)) during the treatment with benralizumab. Data were recorded at baseline and at 6, 12, 24, and 36 months. Mean (SD), n values and regression coefficients (ie, estimate) with 95% CI are reported for each time point.

Similarly, the mean percentage of predicted FEV₁ rose from 75.1% (22.3) at baseline to 83.4% at 6 months, 85.1% at 12 months, and 87.6% at 48 months.

Additional spirometry parameters also improved during treatment. Mean of predicted FVC% increased from 91.9% at baseline to 97.4% at 48 months, while the mean FEV₁/FVC ratio rose from 57.6% to 62.2%. Improvements were also observed in small airway function, the FEF_25–75% predicted increasing from 46.9% (31.7%) to 61.2% at 48 months.

Additionally, fractional exhaled nitric oxide (FeNO), a marker of airway type 2 inflammation, decreased steadily during treatment. Mean FeNO levels dropped from 55.6 ppb (53.9) at baseline to 39.7 ppb at 6 months, and continued to decline to 32.1 ppb at 48 months, consistent with a durable anti-inflammatory effect of benralizumab.

Airway obstruction and small airway function

Fig. 3 illustrates the changes over time in FEV₁/FVC and FEF_25–75% predicted, including subgroup analyses based on baseline impairment. The mean volume of FEV₁/FVC in patients with values ≤ 70 at baseline increased from 49.70 (22.68) at baseline to 65.93 (10.8) at 48 months (Fig. 3A). The percentage of predicted FEF_25–75% in patients with values ≤ 50 at baseline reached 51.23% at 48 months, with a mean increase of 22.32% from baseline (Fig. 3B). Comparisons between obstructed and non-obstructed subgroups for both FEV₁/FVC and FEF_25–75% are shown in Fig. S1.

Fig. 3.

Lung function change in FEV₁/FVC and FEF_25–75% (predicted) during the treatment with benralizumab. The FEV₁/FVC ≤70 patient population (two-dash line) and overall FEV₁/FVC (solid line) are reported in panel A. FEF_25–75 ≤50 patient population (two-dash line) and overall FEF_25–75% (solid line) are in panel B. Mean (SD), n values, and number of patients who changed status from baseline are reported for each time point.

Asthma control and quality of life

Among patients with SEA treated with benralizumab, there were significant and durable improvements in asthma control, as measured by validated patient-reported outcomes (Fig. 4). During benralizumab treatment, asthma control and QoL showed significant improvements, with measurable effects observed as early as 6 months (Fig. 4A). The mean ACT score rose from 15.22 (SD 5.28) at baseline to 21.49 (SD 4.11) at 6 months (Estimate [95% CI] = 6.09 [5.24, 6.94]). The ACT score remained either stable or continued to increase throughout the treatment period, indicating durable and effective control of asthma. In addition, benralizumab enhanced the patients’ QoL as determined by the significant changes of ACQ score over time (mean ACQ: from 6.02 to 1.35, p < 0.0001) (Fig. 4B and C). Moreover, also the AQLQ score improved over time after the start of the treatment with benralizumab (p = 0.0007) (from 4.48 at baseline to 5.91 at 12 months, respectively; (Estimate (95%CI) = 1.31(0.76: 1.87)), reaching a value of 5.78 at 48 months (Estimate (95%CI) = 0.76 (0.02: 1.50)).

Fig. 4.

Change in ACT (A), ACQ (B), and AQLQ (C) scores during the treatment with benralizumab. Data were recorded at baseline and at 6, 12, 24, 36, and 48 months. Mean (SD), n values and regression coefficients (ie, estimate) with 95% CI are reported for each time point.

Among patients with SEA treated with benralizumab, there was marked and durable relief of sinonasal symptoms in those with CRSwNP comorbidity, as shown in Fig. 5. The mean Visual Analogue Scale (VAS) score for nasal symptoms decreased from 4.70 (3.3) at baseline to 3.15 at 6 months, further diminishing to 2.84 at 12 months, and remaining low through the 48-month period (2.50 at 36 months; 2.31 at 48 months). These data indicate a relevant alleviation of upper airway symptom burden.

Fig. 5.

Change in VAS (A) and SNOT 22 (B) score during the treatment with benralizumab. Data were recorded at baseline and at 6, 12, 24, 36 and 48 months. Mean values (SD), n and regression coefficients (ie, estimate) with 95% CI are reported for each time point.

Improvements in patient-reported outcomes were also observed for Sinonasal Outcome Test (SNOT-22) scores, which fell from a mean of 45.9 (21.9) at baseline to 31.2 at 6 months, 28.4 at 12 months, and continued to decline to 22.9 by 48 months. This sustained improvement reflects enhanced quality of life related to sinonasal health.

Oral corticosteroids

Among patients with SEA treated with benralizumab, there was a significant and progressive reduction in oral corticosteroid (OCS) use over the 48-month follow-up period (p < 0.0001), as shown in Fig. 6. The mean dose of OCS constantly decreased from 4.25 mg/day (6.87) at 6 months (−6.23; CI -8.58: −3.89) to a mean dose of 2.42 mg/day (8.58) at 48 months (−8.09; CI -10.46: −5.72).

Fig. 6.

Change in OCS dose in the overall patient population and in patients with baseline OCS dose >5 mg/day during the treatment with benralizumab. Data were recorded at baseline and at 6, 12, 24, and 36 months. Mean (SD), n values, and regression coefficients (ie, estimate) with 95% CI are reported for each time point.

OCS total withdrawal was achieved in 25 of 54 patients (46.3%) within 6 months of starting benralizumab, and in 46 of 52 (88.5%) at 48 months, while 4 patients (6.7%) showed no reduction, maintaining or increasing their baseline dose (Table 2). When stratified by baseline OCS dose (≤5 mg/day or >5 mg/day) (Fig. 6), both the overall population and the >5 mg/day subgroup exhibited comparable patterns of OCS reduction. Table 2 further details the OCS reduction achieved at each time point. By month 6, 79.6% of patients had reduced their OCS dose, with 46.3% achieving complete withdrawal. These proportions increased steadily over time: by month 24, 83.7% of patients had discontinued OCS; by month 48, 88.5% of patients had entirely discontinued OCS. At the final time point, 92.3% of patients had achieved at least a 50% reduction, and only 7.7%continued to be treated with the baseline dose.

Table 2.

Extent of OCS dose reduction achieved by patients during the treatment with benralizumab.

Extent of OCS reduction	Month 6 (n=54)	Month 12 (n=57)	Month 24 (n=49)	Month 36 (n=50)	Month 48 (n=52)
Any reduction	43 (79.63%)	43 (75.44%)	45 (91.84%)	47 (94.00%)	48 (92.31%)
≥90%	25 (46.30%)	36 (63.16%)	41 (83.67%)	43 (86.00%)	46 (88.46%)
≥75%	30 (55.56%)	39 (68.42%)	42 (85.71%)	44 (88.00%)	46 (88.46%)
≥50%	41 (75.93%)	43 (75.44%)	44 (89.80%)	47 (94.00%)	48 (92.31%)
≥25%	42 (77.78%)	43 (75.44%)	45 (91.84%)	47 (94.00%)	48 (92.31%)
No reduction	11 (20.37%)	14 (24.56%)	4 (8.16%)	3 (6.00%)	4 (7.69%)
Interruption	25 (46.30%)	36 (63.16%)	41 (83.67%)	43 (86.00%)	46 (88.46%)

Data are expressed as n (%)

Clinical remission

Fig. 7 shows the number and percentage of patients who achieved any CR, including pCR and cCR, at 6, 12, 24, 36, and 48 months (from the related previous time point) using a FEV₁ cut-off of 150 mL.

Fig. 7.

Number and percentage of patients who achieved and did not achieve CR (either pCR or cCR) at 6, 12, 24, 36, and 48 months during the treatment with benralizumab (from the previous time point). A FEV₁ cut-off of 150 mL was used.

A total of 37 out of 89 patients (60.7%) achieved CR after 6 months; CR increased to 77.6% at 12 months, peaked at 24 months (93.2%), and reached stability at 36 and 48 months (82.0% and 87.9%, respectively). When pCR and cCR were considered separately, the percentage of patients in pCR steadily increased over time from 16.4% at 6 months to 24.1% at 48 months. The percentage of patients achieving cCR saw a peak at 24 months (77.3%) and remained stable thereafter, reaching 70.0% and 63.8% at 36 and 48 months, respectively. Importantly, across all time points, the majority of patients achieved complete rather than partial remission, highlighting not only the efficacy but also the depth of disease control achieved with benralizumab. Patients who did not meet any CR criteria decreased from 39.3% at 6 months to 12.1% at 48 months, further highlighting the long-term efficacy of benralizumab in maintaining remission in a high number of patients. Remission rates assessed from baseline and applying different FEV₁ cut-offs (100 mL and 150 mL) are reported in Fig. S2 and show similar results.

The steady and sustained increase in remission rates over time further demonstrates the long-term disease-modifying potential of benralizumab in patients with severe eosinophilic asthma.

Asthma background medication

Among patients treated with benralizumab, there was a significant reduction in the necessity for background asthma therapies over the 48-month follow-up period, as shown in Table 3.

Table 3.

Change in asthma medication use during the treatment with benralizumab. Descriptive statistics with a summary output of mixed-model on asthma medications other than OCS (ICS, LAMA, and anti-LT) recorded at baseline and during the treatment with benralizumab. Mean (SD) and n values or n (percentage) and the beta regression coefficient (or OR where appropriate) with 95% CI are reported for each time point.

Characteristics	Baseline	Month 6	Month 12	Month 24	Month 36	Month 48	p-value
ICS dose (mcg/day)	982 (401)	884 (451)	895 (429)	828 (397)	784 (406)	638 (356)
	n=110	n=69	n=102	n=91	n=85	n=104
ICS dose							<0.0001a
≥1000	36 (32.7%)	18 (26.1%)	31 (30.1%)	18 (19.8%)	17 (20.0%)	9 (8.7%)
≥500-1000	51 (46.4%)	25 (36.2%)	34 (33.3%)	39 (42.9%)	30 (35.3%)	44 (42.3%)
<500	23 (20.9%)	26 (37.7%)	37 (36.3%)	34 (37.4%)	38 (44.7%)	51 (49.0%)
∗	1	1.10 (0.34 : 3.62)	0.85 (0.31 : 2.28)	3.02 (1.03 : 8.85)	2.71 (0.89 : 8.25)	11.14 (3.10 : 40.09)
#	1	18.24 (18.20 : 18.28)	10.48 (10.45 : 10.50)	10.76 (10.74 : 10.78)	9.02 (9.00 : 9.04)	86.11 (85.92 : 86.30)
Anti LT	53 (42.7%)	37 (45.7%)	40 (35.7%)	27 (27.0%)	30 (30.9%)	19 (18.1%)	<0.0001
	1	0.56 (0.19 : 1.67)	0.27 (0.1 : 0.76)	0.06 (0.02 : 0.21)	0.11 (0.03 : 0.34)	0.02 (0 : 0.07)
	n=124	n=81	n=112	n=100	n=97	n=105
LAMA	86 (74.1%)	58 (77.3%)	72 (68.6%)	54 (58.7%)	53 (55.2%)	50 (49.0%)	<0.0001
	1	0.47 (0.12 : 1.77)	0.25 (0.07 : 0.82)	0.08 (0.02 : 0.31)	0.08 (0.02 : 0.28)	0.05 (0.01 : 0.16)
	n=116	n=75	n=105	n=92	n=96	n=102

^ap value estimated by fitting a multinomial log-linear model corrected for centre and patients variability ∗ ICS dose contrast, ≥500–1000 mcg/day versus >1000 mcg/day; # ICS dose contrast, <500 mcg/day versus >1000 mcg/day

The inhaled corticosteroid (ICS) mean dose (converted to fluticasone equivalents) decreased from 982 μg/day (401) at baseline to 638 μg/day at 48 months (-35%). Patients treated with high-dose ICS (≥1000 μg/day) dropped from 32.7% at baseline to 8.7% at 48 months. Conversely, the proportion of patients on low-dose ICS (<500 μg/day) increased from 20.9% to 49.0% (p < 0.0001). Taken together, these data demonstrate that benralizumab led to a significant reduction in the necessity of corticosteroid therapy over time.

Similar trends were observed for other controller medications. Patients treated with long-acting muscarinic antagonists (LAMA) declined from 74.1% at baseline to 49.0% at month 48 (p < 0.0001). The use of anti-leukotriene (anti-LT) medications also decreased significantly, from 42.7% at baseline to 18.1% at 48 months (p < 0.0001). Regression analysis demonstrated a consistent and significant decline over time in the probability of requiring both LAMA and anti-LT.

Clinical changes by ICS dose over time

Fig. 8 shows changes in key physiological and inflammatory parameters, including pre-bronchodilator FEV₁,% predicted FEV₁, FeNO, FEV₁/FVC, and FEF_25–75% (predicted), stratified by ICS dose changes over time.

Fig. 8.

Change from baseline in pre-bronchodilator FEV₁ (A), percentage of predicted (B), FeNO (C), FEV₁/FVC (D) and FEF 25–75% of predicted (E) stratified by ICS dose change during the treatment with benralizumab. Error bars denote 95%CI. FeNO = fractional exhaled nitric oxide. FEV1 = forced expiratory volume in 1 s. ppb = parts per billion.

Across both ICS dose groups, patients experienced consistent improvements in lung functional parameters. Mean pre-bronchodilator FEV₁ improved by more than 0.4 L at 48 months in both the stable and reduced ICS subgroups. A consistent rise in FEV₁% predicted was observed, with final values aligning across ICS groups regardless of de-escalation. A transient divergence between curves was observed at 12 months. At this time point, patients who maintained a constant ICS dose showed greater improvement in FEV₁ (+0.33 L from baseline) than those who decreased their ICS dose (−0.01 L). Similarly, the increase in FEV₁% predicted was markedly higher in the constant-dose group (+13.64%) than in the ICS-reduction group (+0.38%). This finding likely reflects heterogeneity in the timing and implementation of ICS step-down during the first year of treatment, as some patients may have initiated tapering earlier, when lung function was still improving. In addition, smaller subgroup sizes and inter-individual variability may have amplified differences at this time point. Importantly, this early discrepancy did not persist over time: from 24 months onward, lung-function trajectories converged across ICS categories, supporting that ICS de-escalation during benralizumab treatment did not compromise long-term respiratory outcomes once stable disease control had been achieved.

Airway inflammation, assessed by FeNO, decreased across all subgroups, including those with marked ICS reductions. Notably, patients who tapered from high-to medium- or low-dose ICS showed a FeNO suppression trajectory comparable to those maintained on higher doses, confirming preserved anti-inflammatory efficacy of benralizumab. Predicted FEV₁/FVC ratio and FEF_25–75% also improved across ICS dose categories, with no clinically meaningful differences between patients who maintained or reduced ICS intensity.

The corresponding analyses, including the ICS-increase subgroup, are provided in Fig. S3.

Progressive de-escalation of ICS dose over time

Fig. 9 shows a Sankey plot illustrating the dynamic shifts in inhaled corticosteroid (ICS) dosing categories among patients during the 48-month treatment with benralizumab.

Fig. 9.

Sankey plot of ICS dose during the treatment with benralizumab.

At baseline, 32.7% of patients were on high-dose ICS (≥1000 μg/day), 46.4% on medium-dose (500–1000 μg/day), and 20.9% on low-dose ICS (<500 μg/day). Throughout treatment, a clear trend toward dose reduction emerged. By 48 months, nearly half of the cohort (49.0%) had transitioned to low-dose ICS, while only 8.7% continued the treatment with high doses.

The Sankey flow shows this shift, where a substantial proportion of patients initially treated with high-dose ICS either reduced to medium- or low-dose regimens or discontinued ICS altogether. Similarly, many patients on medium doses at baseline stepped down to low-dose ICS. Only a small portion of patients maintained their baseline therapy, further highlighting the flexibility and effectiveness of benralizumab in supporting ICS de-escalation.

Discussion

This real-world, four-year study in patients with SEA treated with benralizumab provides additional evidence that clinical remission represents a practical, lasting, and important treatment target. Building upon previous findings,¹⁵ this extended follow-up study demonstrates that a substantial portion of patients treated with benralizumab can rapidly achieve and maintain remission over a prolonged period, supporting the importance of early, tailored interventions in biologic-naïve patients, particularly when remission is defined as a multidimensional goal encompassing exacerbation elimination, symptom control, and corticosteroid withdrawal.^2,36 Notably, our study population was characterized by significant baseline airflow obstruction, with nearly three-quarters of patients presenting with an FEV₁/FVC ratio of ≤70% and over 60% exhibiting small-airway dysfunction (SAD), defined as an FEF_25–75% of <50%. Despite this impaired baseline profile, sustained treatment with benralizumab led to marked and progressive improvements in both large and small airway function. By 48 months, the FEV₁/FVC and FEF_25–75% values in these subgroups approached those observed in the overall cohort, indicating functional recovery even in patients with more advanced disease.

Our findings are consistent with XALOC-1, where nearly one-third of patients achieved clinical remission at 96 weeks, and predictors of success included lower baseline OCS use, lower BMI, and higher peak eosinophil counts.^36,37 In our cohort, 75% of patients were biologics-naïve, supporting previous data suggesting that early initiation may improve long-term outcomes.^38,39 Moreover, the current analysis highlights the rapidity and durability of response, and also the progressive adjustment of background therapies over time as a central aspect of remission according to SANI consensus.⁴

The observed reduction in the necessity for OCS maintenance therapy holds significant clinical importance. Long-term systemic corticosteroid administration is associated with substantial morbidity, including increased infection risk, metabolic disturbances, and osteoporosis.⁴⁰ A substantial number of patients in our cohort were able to entirely discontinue OCS supports, supporting the role of benralizumab in steroid-sparing strategies, as also demonstrated by ANANKE⁴¹ and another recent study.⁴² Moreover, reductions in daily ICS doses and simplification of background regimens further reflect disease control and therapeutic optimization, aligning with observations from real-world cohorts 43, 44, 45.

It is also important to underline the alignment between clinical remission and pathophysiological mechanisms. Benralizumab's mechanism of action is based on antibody-dependent cell-mediated cytotoxicity, depleting eosinophils and potentially modulating natural killer (NK) cells' activity.^19,20,46 This mechanism may contribute not only to eosinophil clearance but also to broader immunomodulatory effects, such as improved antiviral defense and reduced exacerbation risk.^47,48

The sustained suppression of eosinophilic inflammation, reflected by near-complete depletion of peripheral eosinophils and a durable reduction in exacerbation rates, supports the hypothesis that benralizumab may modify the course of eosinophilic disease rather than merely provide short-term symptom control. These findings are consistent with long-term observations from both the ANANKE and MELTEMI studies, which demonstrated maintained clinical benefits over extended follow-up periods.^27,41,49

Clinical remission was achieved rapidly and maintained over the entire 48-month follow-up period. More than half of the patients attained complete clinical remission within the first year of treatment, and remission rates remained consistently high thereafter. These findings extend previous real-world observations^15,49 and reinforce the role of benralizumab as an effective long-term strategy for achieving and maintaining remission in severe eosinophilic asthma.

The observed simplification of background therapy is also relevant. Over 4 years, a significant proportion of patients were able to discontinue OCS and reduce or withdraw ICS and other controller medications. This is particularly meaningful considering the well-documented side effects of chronic corticosteroid use^8,50 and supports the paradigm shift toward steroid-sparing treatment strategies in SEA. Although it was not a predefined objective, our data also strongly suggest that sustained remission facilitates de-escalation of maintenance therapy, offering a more favourable long-term risk–benefit balance for patients. These findings align with previous research, demonstrating the safety and efficacy of a structured background therapy de-escalation approach in a similar population.⁴² Our findings also provide real-world evidence on the feasibility of reducing inhaled corticosteroid (ICS) use during effective biologic treatment. In the SHAMAL trial, a substantial proportion of patients receiving benralizumab successfully reduced maintenance ICS/LABA therapy; however, some patients who stepped down to as-needed ICS/formoterol experienced modest declines in lung function and increases in FeNO levels.²⁹ In our cohort, ICS dose reduction occurred progressively over a longer follow-up period and was not associated with deterioration in asthma control, lung function, or airway inflammation. Improvements in FEV₁, FEV₁/FVC, and FEF_25–75% were maintained regardless of ICS dose changes, while FeNO levels continued to decrease over time. The greatest decline in FeNO was observed during the 6 months of treatment and then stabilised. Although FeNO was not included among the SANI remission criteria and was available only in a subset of patients, the sustained reduction observed throughout follow-up provides complementary evidence of durable disease control. Taken together, these findings suggest that a gradual, individualized reduction of ICS therapy may be achievable in patients who attain stable remission with benralizumab, provided that clinical and functional monitoring is maintained.

Over 48 months, the mean pre-bronchodilator FEV₁ increased consistently, along with proportional gains in FEV₁/FVC ratio and FEF₂₅-_75%, suggesting a significant and durable reduction in both large and small airway obstruction. These findings are particularly relevant in the context of PAO and SAD, which are frequent features of SEA and are associated with impaired disease control and faster decline of lung functionality.^10,51 Eosinophilic inflammation, especially in the small airways, plays a key role in SAD and may contribute to fixed obstruction through airway remodeling and mucus plugging..^52,53 By inducing rapid and near-complete eosinophil depletion in blood and tissue, benralizumab has a pivotal immunomodulatory effect^19,20 that likely underpins the observed long-term improvements in lung function, including parameters reflective of distal airway physiology. While short-term benefits in FEV₁ were already documented in phase III trials such as SIROCCO and CALIMA,^23,24 the present findings add real-world evidence of durable recovery of spirometry parameters over 4 years. This supports the role of benralizumab in mitigating progressive airway damage in SEA and emphasizes the importance of early intervention to prevent irreversible structural changes.

The remission framework proposed by the SANI network provided a practical tool for quantifying therapeutic success and aligning clinical goals with patient-centric outcomes. As in our earlier analysis,¹⁵ the SANI criteria proved applicable and sensitive in capturing real-world improvements across a broad range of parameters. Notably, remission was achieved rapidly in a large proportion of patients and was maintained throughout the 48-month study period. Several remission frameworks have been proposed in severe asthma. For example the expert consensus proposed by Menzies-Gow et al.⁶ defines remission through symptom control, absence of exacerbations, OCS independence and optimization of lung function. More recent international initiatives have refined these concepts but continue to differ regarding the role of lung function, biomarker assessment and duration requirements. We selected the SANI criteria because they have been specifically validated in Italian real-world cohorts and allow discrimination between complete and partial remission, facilitating longitudinal assessment of treatment response. Furthermore, using this definition allows to compare this work with other works using the same definition, both for benralizumab and for other boiologics.

Despite the high remission rates observed, approximately 12% of patients did not achieve clinical remission at 48 months. Several factors may contribute to incomplete response, including persistent airflow limitation, advanced airway remodeling, coexisting comorbidities, residual symptoms despite biologic therapy, and the presence of non-type 2 inflammatory mechanisms. Because of the retrospective nature of the study and the limited number of non-remitters, a formal responder/non-responder analysis was beyond the scope of the present investigation. Future studies should further explore predictors of incomplete remission and treatment resistance in severe eosinophilic asthma.

However, this study has some limitations. The primary limitation is its single-arm design without a comparator group, which precludes causal inference and raises the possibility that some of the observed improvements may be attributable in part to regression to the mean. Additionally, the retrospective design may introduce selection bias and limit the granularity of longitudinal data on medication adherence, environmental exposures, and comorbidity management. Nonetheless, the multicentred design, standardized data collection, and consistent follow-up intervals strengthen the validity of these findings.

Conclusion

This four-year real-life study confirms that among patients with SEA treated with benralizumab, a high proportion achieved rapid and stable clinical remission. Importantly, prolonged treatment not only maintains asthma control but also allows de-escalation of background therapies, including OCS and ICS. These findings further demonstrate the role of benralizumab as a disease-modifying agent in SEA and support the use of remission-based targets as a guide for long-term disease management strategies.

Informed consent statement

Informed content was obtained from all participants before enrolment in this study.

Author contributions

Conceptualization, L.P.; Investigation, B.B. (Bianca Beghè), P.C., C.C. (Cristiano Caruso), C.C. (Claudia Crimi)., A.D., G.G., M.L., F.M., B.B. (Benedetta Bondi), Y.E.M., J.G., A.M., M.M., S.P. and M.Z.; Data curation, M.C., A.P. and D.V.; Validation, D.B., F.B. (Fulvio Braido) and A.V.; Writing—original draft, L.P.; Writing—review of the manuscript for intellectual content, G.S., A.S., P.P., F.B. (Francesco Blasi) and G.W.C.; Funding acquisition: SANI Study Group. All authors have read and agreed to the published version of the manuscript.

Disclosure statement

The medical writer used Grammarly (Grammarly (2026). Grammarly Inc. https://www.grammarly.com) exclusively during the finalisation stage of the manuscript to check for minor grammatical and language errors. No generative AI tools were used for content creation, data analysis, or interpretation.

Funding

This work was supported by an unconditional contribution from AstraZeneca S.p.A. Italy.

Conflicts of interest

All authors reported no financial interests or potential conflicts of interest related to this study. L.P. received grants for educational events from AstraZeneca, Chiesi Farmaceutici, Glaxo Smith Kline and speaker fees from AstraZeneca, Chiesi Farmaceutici S.p.A, Glaxo Smith Kline, Guidotti, Grifols, Menarini, Novartis AG; D.B, B.Be, F.Br, D.V. received speaker fees from AstraZeneca, Chiesi Farmaceutici S.p.A, Glaxo Smith Kline, Guidotti, Grifols, Menarini, Novartis AG, Sanofi; P.C. received grants and speaker fees from AstraZeneca and GSK; M.C. received financial grants from AstraZeneca, GSK and Sanofi; C.Cr. received honoraria for lectures from AZ, GSK, Sanofi, Novartis, Resmed, F&P; F.Bl. received financial grants from AstraZeneca, Chiesi Farmaceutici S.p.A and Insmed Inc. and speaker fees from AstraZeneca, Chiesi Farmaceutici S.p.A, Glaxo Smith Kline, Guidotti, Grifols, Insmed Inc., Menarini, Novartis AG, Sanofi-Genzyme, Viatris Inc., Vertex Pharmaceuticals and Zambon; G.W.C. recceived research grants from A. Menarini, Allergy Therapeutics, AstraZeneca, Chiesi Farmaceutici, Faes, Firma, Glaxo Smith Kline, Guidotti-Malesci, Hal Allergy, Innovacaremd, Novartis, OmPharma, RedMaple, Sanofi-Aventis, Sanofi-Genzyme, Stallergenes-Greer, Uriach Pharma, ThermoFisher, Valeas.

Appendix A. Supplementary data

The following is the Supplementary data to this article: