Methods: This multicentre, open-label, single-arm study was done at 138 clinical asthma treatment centres across 17 countries. We enrolled adult patients (age ≥18 years) with severe, eosinophilic asthma (blood eosinophil count ≥150 cells per μL at enrolment or ≥300 cells per μL in the previous year) requiring maintenance oral corticosteroids for at least 3 months preceding enrolment. Patients received benralizumab 30 mg (subcutaneous injection) every 4 weeks for three doses, then every 8 weeks thereafter. The oral corticosteroid reduction phase began at week 4 with daily oral corticosteroid dosages reduced by 1–5 mg every 1–4 weeks depending on the starting dosage, asthma control, and adrenal function status. Adrenal function was assessed with an early morning serum cortisol measurement, followed by adrenocorticotropic hormone stimulation when required, once patients achieved a daily oral corticosteroid dosage of 5 mg/day for 4 weeks. Repeat cortisol measurements were taken for patients with evidence of adrenal insufficiency at first testing. Asthma control was assessed with the Asthma Control Questionnaire-6 (ACQ-6) weekly throughout the induction and oral corticosteroid reduction phases. The primary endpoints were the percentage of patients eliminating daily oral corticosteroids, sustained for at least 4 weeks, and the percentage achieving elimination or a daily prednisone or prednisolone dosage of 5 mg or less, for at least 4 weeks, if the reason for no further reduction was adrenal insufficiency. Safety and efficacy analyses included all patients who received at least one dose of benralizumab and were descriptive. We present results after the oral corticosteroid reduction phase; a maintenance phase is ongoing. The trial is registered with ClinicalTrials.gov, NCT03557307.
Findings: Between April 1, 2018, and Sept 5, 2020, of 705 patients assessed for eligibility, 598 were recruited and all received at least one dose of benralizumab. Overall, 376 (62·88%, 95% CI 58·86–66·76) of 598 patients eliminated oral corticosteroids and 490 (81·94%, 78·62–84·94) of 598 eliminated use or achieved a dosage of 5 mg or less if the reason for stopping the reduction was adrenal insufficiency. Subgroup analysis showed that dosage reductions were achieved irrespective of baseline eosinophil count, baseline oral corticosteroid dosage, or oral corticosteroid treatment duration. Adrenal insufficiency was detected in 321 (60%) of 533 patients at first assessment and in 205 (38%) of 533 patients 2–3 months later. The safety profile was consistent with previous experience. Most patients (448 [75%] of 598) had no asthma exacerbations during the oral corticosteroid reduction phase with an annualised exacerbation rate of 0·63. Of 598 patients, 38 (6%) experienced a total of 46 exacerbations resulting in emergency department or urgent care visits or hospitalisations.
Interpretation: Despite a high prevalence of adrenal insufficiency, most patients with eosinophilic asthma treated with benralizumab achieved elimination of oral corticosteroids or maximal possible reduction using a personalised dosage-reduction algorithm.