ABOVE: © iStock.com, AndrewSoundarajan

Update (August 19): Results of the COVID-OUT trial—which evaluated fluvoxamine, metformin, and ivermectin—were published yesterday in The New England Journal of Medicine. “None of the three medications that were evaluated prevented the occurrence of hypoxemia, an emergency department visit, hospitalization, or death associated with Covid-19,” the authors write in their paper.

Fluvoxamine is a type of selective serotonin reuptake inhibitor currently approved by the US Food and Drug Administration as a treatment for obsessive compulsive disorder, although it is also prescribed for depression. An off-patent drug, fluvoxamine is cheap and widely available, making it potentially attractive as a repurposed COVID-19 treatment—if it can be shown to be effective.

The drug drew splashy headlines last October, when a study reported that it lowered hospitalization risk among COVID-19 outpatients. But debate on this and other fluvoxamine data has been swirling for some months now.

Currently, National Institutes of Health (NIH) guidelines state there is insufficient evidence “to recommend either for or against the use of fluvoxamine for the treatment of COVID-19.” The Infectious Diseases Society of America (IDSA) recommends the drug only be considered for COVID-19 in the context of clinical trials.

Recently, the US Food and Drug Administration (FDA) turned down an emergency use authorization request, filed late last year by University of Minnesota infectious diseases researcher and physician David Boulware, that would have provided official backing for doctors to prescribe the drug as an early treatment for people infected with SARS-CoV-2.

The agency “determined that the data are insufficient to conclude that fluvoxamine may be effective in the treatment of nonhospitalized patients with COVID-19 to prevent progression to severe disease and/or hospitalization,” it announced in a document widely shared online this week.

In the wake of this latest decision, here’s a look at the available evidence on the drug’s efficacy against COVID-19, as well as perspectives of scientists and health experts about what that evidence means.

What do the clinical data on fluvoxamine show?

Despite a recent uptick in media coverage of the drug, most discussions about fluvoxamine’s efficacy in COVID-19 patients draw on data from two randomized controlled trials of unvaccinated adults that were published before this year.

The first, dubbed STOP COVID, was launched as a preliminary study in 2020 by Washington University in St. Louis psychiatrists Angela Reiersen and Eric Lenze. Reiersen has written about how she came up with the idea to test the drug while ill with COVID-19, having read that fluvoxamine could dampen inflammation and improve survival in mice models of sepsis.

Different people have come to very different conclusions about whether or not fluvoxamine should be considered a viable treatment for COVID-19.

The STOP COVID results, published in JAMA in late 2020, hinted that 100 mg of fluvoxamine taken three times a day for 15 days could reduce an infected person’s risk of clinical deterioration, as assessed through a combination of measures including reported shortness of breath and a need for supplemental oxygen or decrease in oxygen saturation.

Limitations including a small number of trial participants and a reliance on patient-completed surveys for data collection meant that “larger randomized trials with more definitive outcome measures” would be needed to determine clinical efficacy, the authors wrote in their paper.

The second published study, which Reiersen and Lenze also worked on, was part of the TOGETHER trial, a multi-arm study designed to evaluate repurposed drug treatments in high-risk outpatients with COVID-19. The fluvoxamine arm included more than 1,400 participants in Brazil, about half of whom took a placebo and half took 100 mg of fluvoxamine two times a day for 10 days.

The results, which sparked the spate of media coverage in October when they were published in The Lancet Global Health, showed that treatment was linked to a statistically significant, moderate reduction in the risk of hospitalization due to COVID-19—provided that “hospitalization” included emergency room visits of at least six hours, a definition the study authors justified based on the way Brazil was operating field hospitals at the time. The drug didn’t show an effect on mortality in the study’s main analysis.

In addition to these studies, there are at least two unpublished datasets from trials that failed to find evidence of fluvoxamine’s benefit as an early COVID-19 treatment.

STOP COVID 2, a follow-up to STOP COVID, tested 100 mg twice a day. This trial was stopped early for futility last year, according to an NIH summary, “after lower than expected case rates and treatment effect were observed.” Reiersen tells The Scientist that the team had trouble recruiting patients and is planning to publish its data soon.

A separate trial run by Boulware and colleagues at Minnesota, known as COVID-OUT, was also recently terminated early for futility. That trial used doses of 50 mg, taken twice daily. Boulware tells The Scientist that the team plans to release the COVID-OUT data when possible.

What are health experts saying about the evidence?

Different people have come to very different conclusions about whether or not fluvoxamine should be considered a viable treatment for COVID-19.

Major US health organizations have taken a relatively consistent line: that there’s not enough evidence to say the drug works. In updates posted in the last few months, the NIH and IDSA note a lack of data on fluvoxamine’s effect on death rates, as well as difficulties interpreting the TOGETHER trial’s definition of hospitalization and differences in adherence to treatment between the fluvoxamine and placebo groups.

The FDA—which, in an unusual move, issued a 27-page summary detailing its reasoning for rejecting Boulware’s request—echoed those concerns in its decision earlier this month. It additionally highlighted the negative findings from COVID-OUT, uncertainty about the drug’s mechanisms of action, and the availability of already approved antiviral drugs such as Pfizer’s paxlovid for US patients.

Various physicians and health experts concur with the view that there’s not enough evidence to recommend fluvoxamine as a COVID-19 treatment. In late April, Abhijit Duggal, vice chair of critical care at the Cleveland Clinic, told MarketWatch: “Based on existing data, this is not something that we feel needs to be looked at as a standard of care for patients with COVID. . . . In the day and age of COVID, there’s a lot of things that have been introduced based on data that would not have really stood the test of scientific rigor in any other situation.”

Justin Morgenstern, an emergency physician in Toronto who has tracked research on potential COVID-19 treatments on his blog First10EM, tells The Scientist he thinks fluvoxamine should not be prescribed for the disease based on available data. (Ontario is one of a couple of regions in North America that have updated their local health guidance to suggest physicians consider fluvoxamine as a treatment in high-risk patients.)

From a statistical perspective, given the huge number of compounds tested against COVID-19 in the last two years, it’s highly likely that some show moderately positive trial results despite no real health benefit, he adds, making replication of findings essential. With as many fluvoxamine trials stopped for futility as trials that found an effect, “there’s a reasonable chance that it’s in that false-positive, or chance-alone finding,” he adds.

The debate around fluvoxamine highlights the need for better and faster ways of organizing and evaluating clinical research, particularly for off-patent drugs that lack the logistical and financial support provided by the pharmaceutical industry.

Others have expressed more enthusiasm about fluvoxamine. University of Wollongong epidemiologist and public health researcher Gideon Meyerowitz-Katz, who has collaborated with the TOGETHER trial’s Edward Mills on an unrelated project, calls the TOGETHER trial a “fantastic piece of research [that] shows a convincing benefit” for fluvoxamine. The study asked its questions in “a robust, scientific way,” he says, adding that some of the negative reaction to it seems unjustified.

Some of the researchers working on fluvoxamine have directly criticized the FDA decision. Boulware, who ran a meta-analysis of existing trial data with Reiersen, Lenze, and others earlier this year, has responded to the agency with a 7-page rebuttal letter that he shared with The Scientist and other news outlets. In it, he details what he says are inconsistencies in how the agency evaluated fluvoxamine compared to other treatments, such as Merck’s molnupiravir, which has an emergency use authorization despite researchers questioning its safety and efficacy.

Boulware tells The Scientist that he is more frustrated by the perceived inconsistency than by the lack of authorization for fluvoxamine, which is less urgently needed in the US now that other treatments are available and most people are vaccinated. Fluvoxamine, which costs just a few dollars per two-week treatment course, could still be useful in countries that don’t have access to expensive treatments, he argues.

Reiersen says that she thinks that fluvoxamine is still a valid treatment option in the US despite the FDA decision, and is working with colleagues to raise awareness about it among physicians through social media, letters to journals, medical education talks, and media interviews. “I do feel that the TOGETHER trial pretty well establishes, even by itself, that fluvoxamine is an effective treatment,” she says.

When will we know more?

Another study that Boulware is involved in, the NIH’s ongoing ACTIV-6 trial, includes a fluvoxamine arm. However, like the COVID-OUT trial, the study is using 50 mg doses, twice a day—an amount that some researchers have suggested may be too low to show a positive effect.

Boulware and Reiersen have said in interviews with The Scientist and other media organizations that negative results for trials using 50 mg could support the idea that 100 mg twice daily is the minimum effective dose. Reiersen notes that “there are lot of things that could contribute to one trial being negative and another being positive.”

Others warn against this way of interpreting negative trial data. “The better interpretation is just that there is no benefit to fluvoxamine, rather than there being a dose-specific benefit,” says Morgenstern. But “it’s certainly a hypothesis”—one that could be tested by using two different dosing regimens in a single trial, he adds.

Data from other ongoing trials may or may not address the question of fluvoxamine’s efficacy directly. Several are studying the drug in combination with other compounds: the TOGETHER trial, for example, has an arm testing fluvoxamine with the inhaled corticosteroid budesonide. The ANTICOV trial in Africa, meanwhile, is investigating a combo therapy of budesonide and fluoxetine, a drug in the same class as fluvoxamine.

Researchers who spoke to The Scientist agree that the debate around fluvoxamine highlights the need for better and faster ways of organizing and evaluating clinical research, particularly for off-patent drugs that lack the logistical and financial support provided by the pharmaceutical industry.

“Pharmaceutical companies have a lot of money to throw behind these things,” Meyerowitz-Katz says. He notes that while fluvoxamine researchers have described struggles getting funding and recruiting enough patients over the last two years, Pfizer was able to run a trial of paxlovid across hundreds of sites with more than 2,200 participants and obtain emergency use authorization—all within the space of a few months last year.

Non-pharma research would benefit from more carefully focusing its resources, Meyerowitz-Katz adds: There have been hundreds of studies on popular drugs such as hydroxychloroquine and ivermectin, which turned out to be ineffective as COVID-19 treatments, but only a few trials done on fluvoxamine.

“If another pandemic comes along, we should try and . . . coordinate research better,” he says, “so that we don’t spend a lot of time asking useless questions, and we spend more of our time asking the important ones.”