The results of this study show that favipiravir did not result in improved outcomes (time to clinical improvement) compared with placebo when used to treat patients with COVID-19 pneumonia. Multiple secondary endpoints (duration of fever, time to discharge or a National Early Warning Score (NEWS) < 3, time until weaning from oxygen therapy, time until weaning from mechanical ventilation, time to hospital discharge) or exploratory endpoints (time to negative SARS-CoV-2 PCR Test, and time to positive SARS-COV-2 antibody IgG Test) did not show differences between groups as well, and additional sensitive analysis performed in the Per Protocol population confirmed the results observed in the mITT population. However, the study did not have enough statistical power to find any differences. In fact, this is an underpowered negative study because of exhaustion of enrolment. Exhaustion of patients’ enrolment over time has been observed in COVID-19 clinical trials. The temporal evolution of the pandemic has taken place in the form of successive waves. This has meant that the trials have had variable enrolment rates over time and that, at the end of the waves, many trials had exhausted their enrolment potential (sites), as has been the case in this study.
Our negative result aligns with previous randomized clinical trials comparing favipiravir with a placebo. Indeed, favipiravir has been shown not to improve outcomes compared with placebo regarding time to virological response [18], time to progression to hospitalization [18, 19], or time to progression to pneumonia [19, 20] when administered to treat early symptomatic COVID-19 infection. Nor has it shown any impact on time to clinical improvement in moderate symptomatic COVID-19 disease [21, 22]. Other trials comparing with a no-treatment arm [23] have also shown no benefit from favipiravir. In a recent trial, favipiravir did not improve clinical outcomes in all patients admitted to hospital with COVID-19. However, patients younger than 60 years seemed to have a beneficial clinical response [30]. Moreover, a meta-analysis including nine studies showed a significant clinical improvement in the favipiravir group versus the control group during seven days after hospitalization. However, supplemental oxygen therapy requiring, transfer to ICU, adverse events and mortality were similar in both groups. Authors concluded that favipiravir possibly exerted no significant beneficial effect in the term of mortality in the general group of patients with mild to moderate COVID‑19 [31].
The tolerability of favipiravir, when used to treat patients with COVID-19 disease, is considered predictable and manageable [27, 30, 31]. In our trial, more adverse effects were detected in the favipiravir group, but these were non-SAEs. In addition, we did not find sufficient statistical evidence to correlate the degree of severity/intensity of the events with the treatment group. Furthermore, laboratory values, vital signs, physical examination, and imaging tests did not reveal any major issues concerning the safety of favipiravir. Two patients died during the study. Both belonged to the favipiravir group. In one of them death was not considered drug related. In the other case, the cause of death was renal failure. The physicians considered that it could be related to the drug, since no other cause was found. However, in a recent study favipiravir seemed a suitable therapeutic option in pediatric patients affected by COVID-19 with kidney injury without a need for dose adjustment [32] and in another study appeared to be well tolerated in adults with renal failure [33].
This study has some strengths and some weaknesses. The endpoints depicted a variety of clinical parameters, including clinical severity and virologic parameters, which led to a comprehensive assessment of the patients. Also, the randomized design was a strength before the initiation of the study. This trial emphasizes the need for randomized controlled trials to show the highest level of evidence after the widespread use in the early times of the pandemic of treatments that have subsequently been shown to be inactive for treating COVID-19 disease. However, the small sample size, has ultimately been the most marked weakness of the study, leading to an underpowered study that precludes firm conclusions.
As a future prospect, several issues must be resolved. A particular problem is the post COVID-19 syndrome, which is troublesome for patients and may last for several weeks or months, leading to long symptomatic periods for the patients and persistent loss of work productivity. Post COVID-19 syndrome is related to an immune response with cytokine release [24] and prolonged pro-coagulant status [25], requiring multidisciplinary management [26] and representing a real challenge for the health system. On the other hand, transitioning from pandemic to endemic, COVID-19 disease is permanently among us. After social distance rules have been relaxed, vaccination programs must be potentiated. For all these reasons, investigation on antivirals is warranted. Whether favipiravir may be more active against COVID-19 when administered for more than ten days, which is the duration of favipiravir treatment in our study, or whether it could be more efficient when administered earlier during the course of the disease, or whether it is more active in patients of Asian ethnicity, or whether it is more effective in younger than 60 years people, are still unanswered questions. It is worth noting that 14 days of treatment has been commonly used in previous clinical trials that have shown a significant clinical improvement on day 14 compared to day 7, as reported in the meta-analysis results [27]. Of note, these results were reported in May 2021. The conclusions supported that favipiravir was a promising agent to treat COVID-19, based on evidence of viral clearance and the aforementioned clinical benefit on day 14. The same positive view was highlighted in a review published in January 2021 (Joshi 2021). The history of favipiravir to treat COVID-19 has evolved from an initial enthusiasm [27, 28] to further disappointment with the availability of new negative data from large randomized clinical trials. As the first studies were conducted in Asian countries, there may be some doubt about whether efficacy is greater in Asian patients. However, this is unlikely, and Japanese investigators have already criticized the widespread use of favipiravir as a compassionate use as too hasty and poorly justified [29].
In conclusion, favipiravir, administered for ten days, did not improve the assessed outcomes compared with a placebo to treat COVID-19 patients with pneumonia admitted to the hospital. These results align with the results from previous randomized trials. However, in the present study the non-serious adverse events were more frequent in the favipiravir group.