Fluvoxamine, a decades-old antidepressant that costs pennies per dose, may reduce the disabling fatigue that defines long COVID for millions of patients. The finding, if confirmed, could give clinicians their first strong evidence-based drug option for a condition that has largely resisted treatment.
A clinical trial tested the drug in Brazilian adults who had experienced persistent fatigue for at least 90 days after a confirmed SARS-CoV-2 infection. Over a 60-day treatment window, fluvoxamine appeared to outperform placebo.
Metformin, the diabetes medication tested in the same trial, produced no meaningful improvement for patients whose long COVID fatigue was already established.
A repurposed drug, a stubborn syndrome
Fluvoxamine, marketed as Luvox, is a selective serotonin reuptake inhibitor originally approved for obsessive-compulsive disorder and depression. It gained attention during the pandemic when smaller studies suggested it might blunt severe acute COVID by dampening inflammatory signaling. Its jump to long COVID fatigue is a different bet: that whatever is driving the persistent exhaustion — immune dysregulation, autonomic dysfunction, or lingering viral fragments — may respond to the same anti-inflammatory pathways the drug appears to modulate.
Long COVID affects millions of people worldwide. Fatigue ranks among its most common and disabling symptoms, keeping patients out of work, school, and daily life for months or years after their initial infection.
The result appears consequential for a patient community that has had few options. According to the trial results, fluvoxamine showed consistent benefits for long COVID fatigue. As an already approved medication with a well-established safety profile, it may have potential for clinical use in treating this condition.
How the trial worked
The trial enrolled adults in Brazil who had documented SARS-CoV-2 infections and had been dealing with fatigue for at least three months. Participants were randomized to fluvoxamine, metformin, or placebo, and followed for 60 days. Fatigue and quality-of-life measures were the primary outcomes.
The trial used a Bayesian adaptive design — a statistical framework that continuously updates the probability of benefit as data accumulate, allowing the trial to stop early once the evidence becomes clear. Traditional trials rely on fixed sample sizes and p-values calculated at the end; Bayesian designs can be faster and, in principle, more ethical when a treatment is either clearly working or clearly not.
The trial authors called the design as consequential as the drug result itself, saying the adaptive approach reached firm conclusions more efficiently than a conventional trial would have.
The reported probability figure is not the same as a classical p-value. It means that, given the trial data, there is a high probability fluvoxamine is more effective than placebo at reducing fatigue — a strong signal by any reasonable standard.
Why fluvoxamine and not metformin?
The divergent results are informative. Metformin has produced encouraging signals in earlier work aimed at preventing long COVID when given during acute infection. But once fatigue is entrenched — sometimes for months — the biology may shift. The data suggest metformin does not reverse that established state, at least not on a 60-day timeline.
Fluvoxamine’s mechanism in long COVID is not fully understood. The drug binds to the sigma-1 receptor, which regulates inflammatory cytokine production, and it also affects platelet activation and serotonin signaling in the gut and brain. Any of these could be relevant to a condition increasingly linked to persistent immune activation and possible viral reservoirs.
Science Blog has previously covered evidence that reactivated latent viruses and immune ghosts from past infections may be driving many long COVID cases — a model that could help explain why an anti-inflammatory antidepressant might help.
Context: what came before
Until now, the drug options for long COVID have been thin. Trials have looked at antivirals, anticoagulants, low-dose naltrexone, and various supplements, with mixed or inconclusive results. The most concrete gains have come from prevention rather than treatment.
A cohort study found that Paxlovid given during acute COVID was associated with fewer COVID-19 hospitalizations and reduced risk of long COVID. That work reinforced antivirals as a prevention tool but said nothing about patients whose symptoms had already set in.
This trial fills that gap. It appears to be the first large randomized trial to show a clinically meaningful benefit for people already sick with long COVID fatigue.
What clinicians will do with this
The practical implication is clear: patients want something they can try today and the finding brings that closer to reality.
Fluvoxamine is off-patent, cheap, and available almost everywhere. That matters enormously for a condition that disproportionately affects working-age adults and that health systems have struggled to address at scale.
But several caveats deserve attention. The trial ran for 60 days; durability of the benefit beyond that window is unknown. The study population was Brazilian adults, and effect sizes may differ in other groups. And fluvoxamine has real side effects — nausea, sleep disturbances, sexual dysfunction, and, in a subset of patients, mood changes. The drug also interacts with other medications through liver enzyme pathways, which means prescribers will need to check for interactions carefully.
The researchers themselves caution that fluvoxamine is not a cure. It reduced fatigue and improved quality of life on average; it did not restore every patient to their pre-COVID baseline. And the trial did not identify which subgroups of long COVID patients benefit most — a critical question given how heterogeneous the syndrome is.
The antidepressant question
The finding will inevitably raise a question patients hear often and resent: is long COVID fatigue partly psychological, and is that why an antidepressant helps?
The trial data do not support that framing. Fluvoxamine’s benefit in COVID contexts has consistently been attributed to its anti-inflammatory and sigma-1 receptor effects, not its mood-lifting properties. Depression scores were not the endpoint; fatigue and quality of life were. And SSRIs vary widely in their anti-inflammatory potency — fluvoxamine is unusual in its sigma-1 affinity, which is why it, and not sertraline or fluoxetine, was tested.
Still, the overlap between neuroinflammation, mood, and fatigue is real. Immune cells in the brain’s protective layer are tied to depression. Whatever fluvoxamine is doing, it is likely acting on shared inflammatory circuitry rather than on mood per se.
What comes next
Replication is the immediate priority. A single positive trial, however well designed, needs confirmation in different populations and with longer follow-up. Researchers will also want to know whether higher doses, longer courses, or combinations with other drugs improve on the current result.
Biomarker work is also overdue. If fluvoxamine helps some patients and not others, identifying who benefits — by inflammatory profile, by symptom cluster, by prior viral load — would let clinicians target the drug rather than prescribe it blind.
For now, the trial offers something long COVID patients have rarely had: a specific, affordable, evidence-backed option to discuss with a physician. That is a modest thing and a significant one at the same time.