Favipiravir for COVID-19 in adults in the community in PRINCIPLE, an open-label, randomised, controlled, adaptive platform trial of short- and longer-term outcomes
Hobbs FR., Gbinigie-Thompson OA., Shanyinde M., Yu LM., Harris V., Dorward J., Hayward G., Saville BR., Berry NS., Evans PH., Thomas NP., Patel MG., Richards D., Hecke OV., Detry MA., Saunders CT., Fitzgerald M., Robinson J., Latimer-Bell C., Allen J., Ogburn E., Grabey J., de Lusignan S., Andersson M., Little P., Butler CC.
Background: Evidence for the effect of favipiravir treatment of acute COVID-19 on recovery, hospital admissions and longer-term outcomes in community settings is limited. Methods: In this multicentre. open-label, multi-arm, adaptive platform randomised controlled trial participants aged ≥18 years in the community with a positive test for SARS-CoV-2 and symptoms lasting ≤14 days were randomised to: usual care; usual care plus favipiravir tablets (loading dose of 3600 mg in divided doses on day one, then 800 mg twice a day for four days); or, usual care plus other interventions. Co-primary endpoints were time to first self-reported recovery and hospitalisation/death related to COVID-19, within 28 days, analysed using Bayesian models. Recovery at six months was the primary longer-term outcome. Trial registration: ISRCTN86534580. Findings: The primary analysis model included 8811 SARS-CoV-2 positive mostly COVID vaccinated participants, randomised to favipiravir (n = 1829), usual care (n = 3256), and other treatments (n = 3726). Time to self-reported recovery was shorter in the favipiravir group than usual care (estimated hazard ratio 1·23 [95% credible interval 1·14 to 1·33]), a reduction of 2·98 days [1·99 to 3·94] from 16 days in median time to self-reported recovery for favipiravir versus usual care alone. COVID-19 related hospitalisations/deaths were similar (estimated odds ratio 0·99 [0·61 to 1·61]; estimated difference 0% [−0·9% to 0·6%]). 14 serious adverse events occurred in the favipiravir group and 4 in usual care. By six months, the proportion feeling fully recovered was 74·9% for favipiravir versus 71·3% for usual care (RR = 1·05, [1·02 to 1·08]). Interpretation: In this open-label trial in a largely vaccinated population with COVID-19 in the community, favipiravir did not reduce hospital admissions, but shortened time to recovery and had a marginal positive impact on long term outcomes.