Inhaled budesonide in the treatment of early COVID-19 (STOIC): a phase 2, open-label, randomised controlled trial

BackgroundMultiple early reports of patients admitted to hospital with COVID-19 showed that chronic respiratory disease were significantly under-represented in these cohorts. We hypothesised the widespread use inhaled glucocorticoids among was responsible for this finding, and tested if would be an effective treatment COVID-19.MethodsWe performed open-label, parallel-group, phase 2, randomised controlled trial (Steroids COVID-19; STOIC) budesonide, compared usual care, adults within 7 days onset mild symptoms. The done community Oxfordshire, UK. Participants randomly assigned budsonide or care stratified age (≤40 years >40 years), sex (male female), number comorbidities (≤1 ≥2). Randomisation using random sequence generation block randomisation a 1:1 ratio. Budesonide dry powder delivered turbohaler at dose 400 μg per actuation. asked take two inhalations twice day until symptom resolution. primary endpoint COVID-19-related urgent visit, including emergency department assessment hospitalisation, analysed both per-protocol intention-to-treat (ITT) populations. secondary outcomes self-reported clinical recovery (symptom resolution), viral symptoms measured Common Cold Questionnare (CCQ) InFLUenza Patient Reported Outcome Questionnaire (FLUPro), body temperature, blood oxygen saturations, SARS-CoV-2 load. stopped after independent statistical review concluded study outcome not change further participant enrolment. This is registered ClinicalTrials.gov, NCT04416399.FindingsFrom July 16 Dec 9, 2020, 167 participants recruited assessed eligibility. 21 did meet eligibility criteria excluded. 146 assigned—73 73 budesonide. For population (n=139), occurred ten (14%) 70 group one (1%) 69 budesonide (difference proportions 0·131, 95% CI 0·043 0·218; p=0·004). ITT population, 11 (15%) (3%) 0·123, 0·033 0·213; p=0·009). needed treat reduce deterioration eight. Clinical 1 shorter (median [95% 6 9] vs 8 [7 11] group; log-rank test p=0·007). mean proportion fever first 14 lower (2%, SD 6) than (8%, 18; Wilcoxon p=0·051) least when group. As-needed antipyretic medication required fewer (27% [IQR 0–50] 50% [15–71]; p=0·025) Fewer had persistent 28 receiving 0·204, 0·075 0·334; p=0·003). total score CCQ FLUPro over better (CCQ difference −0·12, −0·21 −0·02 [p=0·016]; −0·10, −0·00 [p=0·044]). Blood saturations load, by cycle threshold, different between groups. safe, only five (7%) reporting self-limiting adverse events.InterpretationEarly administration reduced likelihood needing medical time COVID-19.FundingNational Institute Health Research Biomedical Centre AstraZeneca. Multiple COVID-19. NCT04416399. From events. Early