Millions of long COVID patients suffer from exhaustion for which no approved treatment exists until now. A large randomized trial offers the first pharmacological approach: the antidepressant fluvoxamine, in psychiatric use for decades, measurably reduces this fatigue. And the effect persists even 30 days after treatment ends.
Results from the REVIVE-TOGETHER trial were published on March 31, 2026 in the Annals of Internal Medicine, one of the world's most prestigious medical journals.
A familiar medication as candidate
Fluvoxamine came to market in the 1990s and its patent has long since expired. In many countries, it is used to treat obsessive-compulsive disorder and severe depression. A month of therapy costs only a few dollars. For long COVID researchers, this makes the drug attractive: if it helps, treatment becomes immediately available globally, without new approval procedures and without prohibitive costs.
What distinguishes fluvoxamine from other antidepressants is its additional action on sigma-1 receptors. These receptors sit at the intersection of immune regulation and cellular stress management. In the laboratory, fluvoxamine inhibits the release of inflammatory messengers through this pathway and reduces cytokine dysregulation. In acute COVID-19 illness, the drug had reduced hospitalization risk in earlier studies. Whether this same mechanism works for chronic long COVID fatigue was the question the REVIVE-TOGETHER trial addressed.
Millions affected worldwide without causal therapy
By the end of 2025, more than 65 million people worldwide had long COVID, according to estimates. The defining symptom: a paralyzing exhaustion that often prevents sufferers from working or performing everyday tasks for years. This fatigue is not ordinary tiredness but a neurological-immunological phenomenon that often worsens with exertion.
Treatment options are limited. Without a therapy that causally addresses fatigue, patients rely on symptom management: physical therapy, adapted activity training, sleep medicine. The research community has intensified focus on long COVID, but concrete therapies are years away.
399 patients, three arms, Brazil
REVIVE-TOGETHER divided 399 adults with persistent fatigue lasting at least 90 days after confirmed SARS-CoV-2 infection into three groups: fluvoxamine (100 mg twice daily), metformin (a diabetes drug with theoretically anti-inflammatory properties), and placebo. Participants came from 22 outpatient clinics in Belo Horizonte and Minas Gerais State. Treatment lasted 60 days, with follow-up through day 90.
Measurement used the Fatigue Severity Scale (FSS), a self-assessment questionnaire with nine items rated one to seven. A score above four is considered clinically meaningful. Healthy adults average 2.3 points.
Results for fluvoxamine are clear: at day 60, FSS improved in the fluvoxamine group versus placebo by 0.43 points. The statistical certainty that fluvoxamine beats placebo is 99 percent. More remarkable still: at day 90, 30 days after treatment ended, the difference versus placebo had widened to 0.58 points. The positive effect actually strengthened after therapy stopped. Quality of life also improved across multiple dimensions.
Metformin showed no measurable benefit. This result highlights the specificity of fluvoxamine's effect: not every anti-inflammatory substance helps long COVID fatigue.
What the study does not explain
Results come with an important caveat: the trial did not measure biological markers, blood inflammation levels, or immune parameters. Researchers cannot say whether fluvoxamine actually addresses the causes of long COVID fatigue or whether improvement occurs through other paths. The authors themselves note that observed benefits might reflect improvements in general wellbeing and symptom perception rather than modification of the underlying disease biology.
Additionally, all participants came from Brazil. Whether results translate to European patients, exposed to different infection waves and often carrying different comorbidities, remains open. Researchers also note that a related medication, fluoxetine, did not reduce exhaustion in studies of patients with myalgic encephalomyelitis and chronic fatigue syndrome. Not all fatigue disorders respond equally to serotonin-active substances.
In comparison: How other chronic infections found therapy
How long the path can be from first positive trial to established therapy appears in histories of other chronic infections. Hepatitis C was hard to treat until the early 2000s: interferon-based therapies helped only some patients and carried substantial side effects. Beginning in 2014, direct antiviral agents arrived offering cure rates above 95 percent, orally, in weeks, with few side effects. From first efficacy signals to clinical routine spanned roughly ten years.
HIV followed a similar arc: in 1996, highly active antiretroviral combination therapy showed that viral load could be durably suppressed. Today, HIV-positive people in developed countries with guideline-concordant therapy have lifespans nearly equal to the general population. The path was not linear. It required many follow-up trials, stepwise refinements, and substantial political pressure.
For long COVID, REVIVE-TOGETHER is not an endpoint but a starting point.
Three conditions before fluvoxamine reaches clinical practice
For the drug to actually reach routine long COVID care, at least three conditions must be met.
First: replication in other populations. One Brazilian trial is a signal, not proof. Studies in Europe and North America must show whether the effect holds stable.
Second: biomarker analysis. Without biological correlates, neither explaining why fluvoxamine works nor predicting which patients benefit becomes possible. Clinics prescribing the drug uncritically risk treating patients for whom it is unsuitable.
Third: medical oversight. Fluvoxamine has drug interactions and discontinuation symptoms requiring physician supervision. In the trial, adverse event rates in the fluvoxamine group were 20 percent, surprisingly lower than the placebo group at 29.7 percent. This result itself is not yet understood and shows how much research still has to clarify.
Patients should not take the drug on their own. Fluvoxamine currently lacks approval for any long COVID indication. What the trial has accomplished: it has set the first convincing pharmacological signal and thereby prepared the ground for the next research phase.