Editorial 

January 12, 2023

Lack of Benefit of Fluvoxamine for COVID-19

Adarsh Bhimraj, Jason C. Gallagher

JAMA. Published online January 12, 2023. doi:10.1001/jama.2022.23954

The COVID-19 pandemic caught society unprepared. The lack of therapeutic agents for coronaviruses led to a scramble to investigate approved medications for repurposing against SARS-CoV-2.¹ Successes were achieved with immunomodulatory agents for the inflammatory disease processes of COVID-19, but the antiviral search has been less productive. One of the more interesting therapeutic candidates investigated is fluvoxamine.

Fluvoxamine is an inexpensive generic antidepressant, a selective serotonin reuptake inhibitor, with putative anti–SARS-CoV-2, anti-inflammatory, and antiplatelet activity.² Over the course of the pandemic, there have been multiple clinical trials evaluating the efficacy and safety of fluvoxamine for the treatment of mild to moderate COVID-19. At a time when health care personnel and resources were scarce, it is remarkable that these outpatient trials were conducted quickly with adaptive and contactless designs to answer critical questions about a new disease with uncertainties regarding management.

In this issue of JAMA, McCarthy and colleagues³ report results from the Accelerating COVID-19 Therapeutic Interventions and Vaccines (ACTIV-6) platform randomized clinical trial that was designed to test outpatient treatments for mild to moderate COVID-19 and was conducted in the US during the periods when the Delta variant and early Omicron subvariants dominated. Among the 1288 participants who completed the trial, 674 were randomized to receive 50 mg of fluvoxamine twice daily for 10 days and 614 were randomized to placebo. The lower dose of fluvoxamine (50 mg twice daily) was chosen for better tolerability and adherence compared with a dose of 100 mg twice daily or 3 times daily used in earlier COVID-19 trials.

Although not a requirement for enrollment, high-risk comorbidities (36.4% had a body mass index >30, 24.4% had hypertension, 13.2% had asthma, and 9.2% had diabetes) were frequent and 67% received at least 2 doses of a SARS-CoV-2 vaccine. Few participants (4%) were treated concomitantly with other COVID-19 treatments approved or authorized for emergency use by the US Food and Drug Administration. The median time to recovery was 12 days in the fluvoxamine group compared with 13 days in the placebo group, and there was no significant benefit for the primary outcome of time to sustained recovery. For the secondary composite outcome of hospitalization, urgent care visit, emergency department visit, or death through day 28, the fluvoxamine group had an event rate of 3.9% compared with the event rate of 3.8% in the placebo group. Hospitalization and mortality were extremely low in this study, with 1 participant in the fluvoxamine group and 2 participants in the placebo group being hospitalized and no deaths in either group.

This study failed to demonstrate that fluvoxamine is effective for the treatment of mild to moderate COVID-19. How should we interpret the results of the ACTIV-6 trial³ in the context of evidence from earlier trials on fluvoxamine? A more practical question for frontline clinicians is: Does fluvoxamine have a role in the treatment of mild to moderate COVID-19 at this phase of the pandemic?

Earlier trials that evaluated fluvoxamine for the treatment of COVID-19 showed variable results. The STOP COVID and TOGETHER trials reported benefits, but the STOP COVID 2 and COVID-OUT trials did not.^4-7 As with most studies of COVID-19 therapeutics, interpretation of these trials was complicated by their differing locations, predominant circulating variants, time periods, baseline risks of progression in their populations, and vaccination rates. They also used differing dosing regimens and evaluated different clinical outcomes.

The STOP COVID trial⁷ was one of the earliest placebo-controlled trials conducted in the US with participants randomized to 100 mg of fluvoxamine up to 3 times daily for 15 days (n = 80) or placebo (n = 72). The primary outcome was clinical deterioration defined by a composite of dyspnea, hospitalization for dyspnea or pneumonia, and having an oxygen saturation as measured by pulse oximetry (Spo₂) less than 92% on room air. Patients treated with fluvoxamine had a statistically significant lower likelihood of clinical deterioration (0%) compared with placebo (8.3%). Even though a benefit was shown, the sample size was small, and the low event rate made the finding extremely fragile.

A larger placebo-controlled, randomized, platform trial,⁶ TOGETHER was conducted in Brazil among primarily unvaccinated high-risk adults with mild to moderate COVID-19. Participants were assigned to 100 mg of fluvoxamine twice daily for 10 days (n = 741) or placebo (n = 756). A lower proportion of patients in the fluvoxamine group (11%) compared with the placebo group (16%) experienced the composite outcome of either requiring observation in an emergency setting for longer than 6 hours or admission to a tertiary hospital. In this trial, extended observation was used as a proxy for hospitalization because hospitals were at capacity in Brazil during this phase of the pandemic. There were no significant differences between fluvoxamine and placebo for the secondary outcomes of viral clearance, all-cause hospitalization, time to hospitalization, number of days in the hospital, time to recovery, days of mechanical ventilation, or mortality.

The STOP COVID 2 trial⁴ used the same outcomes as the STOP COVID trial,⁷ but was halted for futility in May 2021 due to low enrollment and event rates.⁸ In the COVID-OUT trial,⁵ participants were randomized to 50 mg of fluvoxamine twice daily for 14 days (n = 334) or placebo (n = 327), and the primary outcome was progression to a composite of Spo₂ of 93% or less on room air, emergency department visit, hospitalization, or death. Fluvoxamine did not show a significant effect in preventing disease progression with an adjusted odds ratio of 0.94 (95% CI, 0.66 to 1.36) for that composite outcome.⁵

Before trying to reconcile the heterogenous results of these studies, we should take measure of the utility of the outcomes measured and their meaning to patients and clinicians in the current state of the pandemic. Discomfort, disability, and death are more patient-centered outcomes than those that measure disease characteristics or health care use. Hospitalization and proxies for hospitalization have been a common primary outcome reported in many outpatient COVID-19 treatment trials because of the ease of measurement, objectivity, and higher event rates found earlier in the pandemic.

In the TOGETHER trial,⁶ the primary outcome was a composite of hospitalization and extended time in an emergency setting. Using outcomes of health care resource use such as hospitalizations and emergency department visits alone without a concomitant measure of patient discomfort or disability has limitations. Hospitalization or location of care can be influenced by institutional bed availability and individual caregiver practices. When hospitals were at crisis capacity in some locations, patients with severe COVID-19 and hypoxemia were being managed at home, whereas in other settings patients with mild to moderate COVID-19 were admitted to the hospital due to fear of complications.

Outcomes that capture patient discomfort or disability coupled with degree of respiratory dysfunction as reported in the STOP COVID and STOP COVID 2 trials^4,7 or measures of duration and severity of symptoms as reported in the ACTIV-6 trial³ are probably more meaningful. Although the use of composite outcomes improves statistical power and decreases trial duration and cost, combining heterogenous outcomes is problematic. For mild to moderate COVID-19, the individual clinical outcome of time to improvement of symptoms reported in the ACTIV-6 trial³ is a meaningful outcome to patients in addition to the outcomes of progression to severe disease or death. The differences in the results between the studies could be due to how the outcomes were defined, measured, or combined.

Another explanation could be due to the change in severity of COVID-19 over time. COVID-19 has evolved to become a less severe disease, likely due to both changes in population immunity and circulating variants and could be a potential explanation why more recent studies like the ACTIV-6 trial³ did not show a benefit with use of fluvoxamine. The between-trial difference in the dosing regimens is a possible but likely less influential contributor. Despite the difficulties in appraising studies across the pandemic, the totality of evidence for fluvoxamine does not support its current use for treatment of mild to moderate COVID-19. This is especially so as antiviral treatment alternatives like ritonavir-boosted nirmatrelvir, remdesivir, and molnupiravir with stronger supporting evidence are now available.

The researchers of the fluvoxamine studies deserve credit for their persistence in investigating the possible role of this inexpensive therapy. In a way, the repeated study of fluvoxamine over the course of the COVID-19 pandemic brings up another issue—the need for continued study of authorized antivirals as the pandemic changes. Drugs shown to be effective earlier in immune-naive populations and against more problematic SARS-CoV-2 variants require further study to define their role in the current landscape of COVID-19.

At a time when hospitalization for COVID-19 is less common, do outpatient therapies shown to prevent it have other beneficial clinical effects? Can these treatments cause a meaningful reduction in severity or duration of symptoms? Can they decrease disability and improve quality of life in the short-term and in the long-term during the post–COVID-19 condition (long COVID)? It is critical that future clinical trials and observational studies adapt to the evolving pandemic and evaluate such patient-centered outcomes.