Pavord 2012a

Methods	Multicentre, double‐blind, placebo‐controlled trial
Participants	621 participants with severe asthma despite receiving high doses of standard asthma medications Main inclusion/exclusion criteria: ≥ 3% sputum eosinophils or blood eosinophil ≥ 300 cells/μL ≥ 2 exacerbations in previous 12 months maintenance treatment with high‐dose ICS (i.e. ≥ 880 μg/d FP or equivalent daily); + additional controller; ± maintenance OCS Age mean (SD) years: mepolizumab 750 mg, 48.6 (11.1); mepolizumab 250 mg, 49 (11.6); mepolizumab 75 mg, 50.2 (10.8); placebo, 46.4 (11.3) Males n (%): mepolizumab 750 mg, 93 (60%); mepolizumab 250 mg, 93 (61%); mepolizumab 75 mg, 104 (68%); placebo, 97 (63%) Baseline mean (SD) FEV1 % predicted: mepolizumab 750 mg, 61% (16); mepolizumab 250 mg, 59% (17); mepolizumab 75 mg, 60% (16); placebo, 59% (15) Allocation: mepolizumab 750 mg, 156; mepolizumab 250 mg, 152; mepolizumab 75 mg, 154; placebo, 159
Interventions	13 total intravenous infusions of mepolizumab (750 mg), mepolizumab (250 mg), mepolizumab (75 mg) or placebo given every 4 weeks
Outcomes	Primary outcomes Frequency of clinically significant exacerbations of asthma Secondary outcomes Time to first clinically significant exacerbation requiring oral or systemic corticosteroids, hospitalisation, and/or ED visits Frequency of exacerbations requiring hospitalisation (including intubation and admittance to an ICU) or ED visits Time to first exacerbation requiring hospitalisation or ED visit Frequency of investigator‐defined exacerbations Time to first investigator‐defined exacerbation Mean change from baseline in clinic pre‐bronchodilator FEV1 over the 52‐week treatment period Mean change from baseline in clinic post‐bronchodilator FEV1 over the 52‐week treatment period Mean change from baseline in ACQ score
Notes	52‐week study conducted at 81 centres in 13 countries (Argentina, Australia, Canada, Chile, France, Germany, South Korea, Poland, Romania, Russia, Ukraine, the UK and the USA) Supported by GlaxoSmithKline
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Central telephone‐based system and computer‐generated randomly permuted block schedule stratified by whether treatment with OCS was required
Allocation concealment (selection bias)	Low risk	Mepolizumab and placebo were prepared by unmasked site staff who were not involved in study assessments
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Mepolizumab and placebo were prepared by unmasked site staff who were not involved in study assessments. Both treatments were identical in appearance and were given to participants by a masked member of the site staff
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Data analysts were masked to treatment allocation
Incomplete outcome data (attrition bias) All outcomes	Low risk	All participants accounted for with information on reasons for having withdrawn. Some participants not included in results due to ‘poor efficacy’
Selective reporting (reporting bias)	Low risk	No apparent indication of reporting bias

ACQ: Asthma Control Questionnaire; ALT: alanine aminotransferase; Alk Phos: alkaline phosphatase; AQLQ: Asthma Quality of Life Questionnaire; AST: aspartate aminotransferase; ECP: eosinophil cationic protein; ED: emergency department; FeNO: exhaled fraction of nitric oxide; FEV₁ : Forced expiratory volume in 1 second; FP: fluticasone propionate; FVC: forced vital capacity; HRQoL: health‐related quality of life; ICS: inhaled corticosteroid; ICU: intensive care unit; IL: interleukin; IQR: interquartile range; IV: intravenous; JACQ: Juniper Asthma Control Questionnaire; OCS: oral corticosteroids; PC₂₀ : histamine provocative concentration causing a 20% drop in FEV₁;PEFR: peak expiratory flow rate; SC: subcutaneous; SD: standard deviation; SGRQ: St. George's Respiratory Questionnaire; ULN: Upper Limit of Normal; VC: vital capacity. ^aQTc(F): a measure of the time between the start of the Q wave and the end of the T wave in the heart's electrical cycle, corrected for the heart rate using Fredericia's formula.