Efficacy and Safety of Favipiravir in Moderate COVID-19 Pneumonia Patients without Oxygen Therapy: A Randomized, Phase III Clinical Trial

Abstract

Introduction: Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), the cause of coronavirus disease 2019 (COVID-19), is an enveloped, single-stranded RNA virus. Favipiravir is an orally administrable antiviral drug whose mechanism of action is to selectively inhibit RNA-dependent RNA polymerase. A preliminary trial in COVID-19 patients reported significant improvements across a multitude of clinical parameters, but these findings have not been confirmed in an adequate well-controlled trial. We conducted a randomized, single-blind, placebo-controlled Phase III trial assessing the efficacy and safety of favipiravir in patients with moderate pneumonia not requiring oxygen therapy.

Methods: COVID-19 patients with moderate pneumonia (SpO₂ ≥ 94%) within 10 days of onset of fever (temperature ≥ 37.5 °C) were assigned to receive either placebo or favipiravir (1800 mg twice a day on Day 1, followed by 800 mg twice a day for up to 13 days) in a ratio of 1:2. An adaptive design was used to re-estimate the sample size. The primary endpoint was a composite outcome defined as the time to improvement in temperature, oxygen saturation levels (SpO₂), and findings on chest imaging, and recovery to SARS-CoV-2-negative. This endpoint was re-examined by the Central Committee under blinded conditions.

Results: A total of 156 patients were randomized. The median time of the primary endpoint was 11.9 days in the favipiravir group and 14.7 days in the placebo group, with a significant difference (p = 0.0136). Favipiravir-treated patients with known risk factors such as obesity or coexisting conditions provided better effects. Furthermore, patients with early-onset in the favipiravir group showed higher odds ratio. No deaths were documented. Although adverse events in the favipiravir group were predominantly transient, the incidence was significantly higher.

Conclusions: The results suggested favipiravir may be one of options for moderate COVID-19 pneumonia treatment. However, the risk of adverse events, including hyperuricemia, should be carefully considered.

Trial registration: Clinicaltrials.jp number: JapicCTI-205238.

Keywords: COVID-19; Favipiravir; Moderate pneumonia not requiring oxygen therapy; Oral antiviral agent; Phase III clinical trial; SARS-CoV-2; Treatment efficacy.

Fig. 1

Of the 156 patients who underwent randomization, 119 (85 favipiravir, 34 placebo) completed the assigned study drug treatment. The study drug was discontinued in 22 patients in the favipiravir group and 15 in the placebo group, and, among these patients, the scheduled observations were continued to Day 28 in 15 patients in the favipiravir group and 7 in the placebo group. The study was discontinued for 4 patients in the favipiravir group and 1 in the placebo group. All 156 patients (107 favipiravir, 49 placebo) were included in the efficacy analysis as modified ITT population

Fig. 2

Kaplan–Meier curves for improvement in COVID-19 symptoms (i.e., temperature, SpO2, and findings on chest imaging, and SARS-CoV-2-negative conversion on qualitative testing) for 28 days defined as the primary end point among all patients in the trial (primary analysis) (Panel a) and two sub-groups according to whether the patients had IgA or IgG antibodies (Panel b) or not (Panel c) at randomization. The red curve represents the favipiravir group, and the blue curve represents the placebo group. The p value for the primary analysis (Panel a) was calculated by the log-rank test based on the weighted Z statistic. The others (Panels b and c) were calculated by a log-rank test. Hazard ratios (HR) for favipiravir to placebo were calculated using the Cox proportional hazards model adjusted for sex and age at baseline

Fig. 2

Kaplan–Meier curves for improvement in COVID-19 symptoms (i.e., temperature, SpO2, and findings on chest imaging, and SARS-CoV-2-negative conversion on qualitative testing) for 28 days defined as the primary end point among all patients in the trial (primary analysis) (Panel a) and two sub-groups according to whether the patients had IgA or IgG antibodies (Panel b) or not (Panel c) at randomization. The red curve represents the favipiravir group, and the blue curve represents the placebo group. The p value for the primary analysis (Panel a) was calculated by the log-rank test based on the weighted Z statistic. The others (Panels b and c) were calculated by a log-rank test. Hazard ratios (HR) for favipiravir to placebo were calculated using the Cox proportional hazards model adjusted for sex and age at baseline

Fig. 3

Individual symptom specific hazard ratios using the Cox proportional hazards model adjusted for sex, age, the categorical variables that correspond to individual symptoms, and their interactions. Hazard ratios are plotted as black diamonds. Horizontal lines represent 95% confidence intervals

Fig. 4

Time course of cumulative improvement rates on chest X-ray imaging. An asterisk on the difference in improvement rate indicates that there is a significant difference

Fig. 5

Time course of patient status in accordance with the 7-point ordinal scale. Any missing clinical status score after the day of discharge was imputed by status score 1 (not hospitalized, no limitations on activities); other missing scores were imputed by last observation carried forward. Each number means the following clinical status; 1 not hospitalized, no limitations on activities; 2 not hospitalized, limitation on activities; 3 hospitalized, not requiring supplemental oxygen; 4 hospitalized, requiring supplemental oxygen; 5 hospitalized, on non-invasive ventilation or high-flow oxygen devices; 6 hospitalized, on invasive mechanical ventilation or extracorporeal membrane oxygenation; 7 death

Fig. 6

Sub-group specific hazard ratios using the Cox proportional hazards model adjusted for sex, age, the categorical variables that correspond to sub-groups, and their interactions. Hazard ratios are plotted as black diamonds. Horizontal lines represent 95% confidence ntervals. Nine and five patients in the favipiravir and placebo groups, respectively, were censored on Day 0 because they were SARS-CoV-2-negative before treatment initiation. Coexisting conditions included prognostic factors such as chronic respiratory illness, chronic renal disease, diabetes mellitus, hypertension, and cardiovascular illness