ABSTRACTAimTo assess the efficacy and safety of favipiravir in adults with moderate to severe coronavirus disease 2019 (COVID-19).MethodsIn this randomized, double-blind, multicenter, phase 3 trial, adults (21-80 years) with real-time reverse transcriptase polymerase chain reaction (rRT-PCR) confirmed SARS-CoV-2 infection and presenting with moderate to severe COVID-19 and requiring hospitalization were randomized 1:1 to oral favipiravir (day 1: 1800 mg BID and days 2-10: 800 mg BID) (FPV) plus standard supportive care (SoC) versus placebo plus SoC (placebo). The primary endpoint was time to resolution of hypoxia.ResultsIn total, 353 patients were randomized to receive either FPV or placebo (175 and 178 in the FPV and placebo groups, respectively). Overall, 76% of the patients (240/315, 78% in FPV vs. 75% in placebo group) reached resolution of hypoxia on or before day 28. The median time to resolution of hypoxia was 7 days in the FPV group and 8 days in the placebo group. Treatment effect was not significant [Hazard ratio (HR) (95% CI): 0.991 (0.767, 1.280) (p=0.94)].Patients in the lower NEWS-2 clinical risk subgroup were more likely to achieve shorter time to resolution of hypoxia with the median time to resolution of hypoxia of 6 days in FPV and 7 days in placebo group [HR (95% CI): 1.21 (0.847, 1.731) (p=0.29)]; shorter time to hospital discharge with a median time to discharge of 8 and 10 days in the FPV and placebo group, respectively [HR (95% CI): 1.47 (1.081, 1.997) (p=0.014)]; and shorter time to improvement by 1-point improvement over baseline in WHO 10-point clinical status score with the median time to improvement by 1-point from baseline of 6 and 7 days in the FPV and placebo group, respectively [HR (95% CI): 1.16 (0.830, 1.624) (p=0.38)] than higher NEWS-2 clinical risk subgroup.Treatment emergent adverse event (TEAEs) were experienced by 62/334 (19%) patients [35/168 (21%) patients in FPV and 27/166 (16%) in placebo group]. Hyperuricaemia/increased blood uric acid was reported in 9 (3%)/2 (1%) patients [8 (5%)/1(1%) patients in FPV and 1 (1%)/1(1%) in placebo group], which were of mild intensity and transient. Overall, 36 serious adverse events (SAEs) were reported, 20 in FPV and 16 in placebo group.ConclusionThe trial did not find favipiravir to be effective in moderate to severe, hospitalized COVID-19 patients; favourable clinical trends were observed in patients with lower NEWS-2 risk when early administration of favipiravir could be achieved.