BACKGROUND:ANDHI was done to assess the efficacy of benralizumab, including onset of effect and impact on health-related quality of life (HRQOL), exacerbation rate, lung function, and nasal polyposis symptoms. METHODS:This phase 3b, randomised, double-blind, parallel-group, placebo-controlled ANDHI study was completed in adults (aged 18-75 years) with severe eosinophilic asthma with at least 2 exacerbations in the previous year, despite high-dose inhaled corticosteroid plus additional controllers, screening blood eosinophil counts of at least 150 cells per μL, and an Asthma Control Questionnaire 6 (ACQ-6) score of 1·5 or more. Patients who met eligibility criteria were randomly assigned (2:1; stratified by previous exacerbation count [two, or three or more], maintenance oral corticosteroid use, and region), using an integrated web-based response system, to receive benralizumab at 30 mg every 8 weeks (first three doses given 4 weeks apart) or matched placebo for 24 weeks. Primary efficacy measure was annualised asthma exacerbation rate, with rate ratio (RR) calculated over the approximate 24-week follow-up. Secondary efficacy measures included change from baseline to end of treatment (week 24) in St George's Respiratory Questionnaire (SGRQ) total score (key secondary endpoint), FEV1, peak expiratory flow (PEF), ACQ-6, Predominant Symptom and Impairment Assessment (PSIA), Clinician Global Impression of Change (CGI-C), Patient Global Impression of Change (PGI-C), and Sino-Nasal Outcome Test-22 (SNOT-22). All efficacy analyses, except for SNOT-22, were summarised and analysed using the full analysis set on an intention-to-treat population (all randomly assigned patients receiving investigational product, regardless of protocol adherence or continued participation in the study). SNOT-22 was summarised for the subgroup of patients with physician-diagnosed nasal polyposis with informed consent. This study is registered with ClinicalTrials.gov, NCT03170271. FINDINGS:Between July 7, 2017, and Sept 25, 2019, 656 patients received benralizumab (n=427) or placebo (n=229). Baseline characteristics were consistent with severe eosinophilic asthma. Benralizumab significantly reduced exacerbation risk by 49% compared with placebo (RR estimate 0·51, 95% CI 0·39-0·65; p<0·0001) over the 24-week treatment period and provided clinically meaningful and statistically significant improvement from baseline to week 24 in SGRQ total score versus placebo (least squares mean change from baseline -8·11 (95% CI -11·41 to -4·82; p<0·0001), with similar differences at earlier timepoints. Benralizumab improved FEV1, PEF, ACQ-6, CGI-C, PGI-C, PSIA, and SNOT-22 at week 24 versus placebo, with differences observed early (within weeks 1 to 4). Adverse events were reported for 271 (63%) of 427 patients on benralizumab versus 143 (62%) of 229 patients on placebo. The most commonly reported adverse events for the 427 patients receiving benralizumab (frequency >5%) were nasopharyngitis (30 [7%]), headache (37 [9%]), sinusitis (28 [7%]), bronchitis (22 [5%]), and pyrexia (26 [6%]). Fewer serious adverse events were reported for benralizumab (23 [5%]) versus placebo (25 [11%]), and the only common serious adverse event (experienced by >1% of patients) was worsening of asthma, which was reported for nine (2%) patients in the benralizumab group and nine (4%) patients in the placebo group. INTERPRETATION:Our results extend the efficacy profile of benralizumab for patients with severe eosinophilic asthma, showing early clinical benefits in patient-reported outcomes, HRQOL, lung function, and nasal polyposis symptoms. FUNDING:AstraZeneca.