Editor's Note: Metformin, often hailed as a "miracle drug" for its wide range of therapeutic effects, was also considered a potential treatment for COVID-19 during the pandemic. Unfortunately, the large-scale COVID-OUT clinical trial, published in NEJM in 2022, did not meet its primary endpoint. However, after excluding patients with "oxygen saturation ≤93%," metformin showed a trend toward reducing the risk of emergency visits, hospitalization, and death, while also displaying potential benefits in reducing long COVID. Recently, an analysis of viral load from this trial was published in Clinical Infectious Diseases (CID), the official journal of the Infectious Diseases Society of America (IDSA). This analysis demonstrated that metformin can significantly lower the viral load in outpatient COVID-19 patients, reinforcing the drug's antiviral properties and explaining its observed benefits in preventing severe COVID-19 and reducing long COVID. Currently, this article ranks as the most-read on the CID website.Metformin has demonstrated antiviral activity against RNA viruses, including SARS-CoV-2. The drug’s mechanism appears to involve targeting the host’s mTOR pathway to inhibit protein translation.
The COVID-OUT trial was a 2×3 randomized, placebo-controlled, double-blind study that aimed to evaluate the efficacy and safety of metformin, fluvoxamine (a selective serotonin reuptake inhibitor), and ivermectin (a broad-spectrum antiparasitic) in outpatient COVID-19 patients. Previous reports have shown that metformin reduced the 28-day risk of hospitalization or death by 58%, reduced the 14-day risk of emergency visits, hospitalization, or death by 42%, and lowered the risk of long COVID by 42% after 10 months.
The recent CID publication involved 999 patients who self-collected nasal swabs on day 1 (n=945), day 5 (n=871), and day 10 (n=775). Viral load was measured using RT-qPCR.
Key Findings:
- Compared to placebo, metformin reduced SARS-CoV-2 viral load by an average of 3.6-fold (-0.56 log10 copies/mL; 95% CI: -1.05 to -0.06; P=0.027).
- On days 5 and 10, patients in the metformin group were less likely to have detectable viral loads than those in the placebo group (OR 0.72; 95% CI: 0.55 to 0.94).
- The rate of viral rebound (defined as higher viral load on day 10 than day 5) was significantly lower in the metformin group compared to the placebo group (3.28% vs 5.95%; OR 0.68; 95% CI: 0.36 to 1.29).
- Metformin’s effects were consistent across various subgroups and increased over time.
- Neither ivermectin nor fluvoxamine showed effects superior to placebo.
How Does Metformin Work Against COVID-19? The drug exerts both antiviral and anti-inflammatory effects. The antiviral mechanism mainly involves inhibiting mTOR, a key factor in protein translation. Since protein synthesis is crucial for SARS-CoV-2 replication, metformin may suppress viral replication through this pathway. In vitro studies have shown that metformin exhibits antiviral activity against Zika virus and hepatitis C virus, also through mTOR inhibition.
Additionally, metformin has multiple anti-inflammatory mechanisms. It inhibits the production of cytokines such as IL-1, IL-6, and TNF-α, which are linked to the severity of COVID-19. Animal studies have demonstrated that metformin can inhibit inflammasome activation, IL-1 production, and IL-6 secretion, while increasing anti-inflammatory cytokine IL-10, thereby reducing lung injury. It also suppresses mitochondrial ATP and DNA synthesis, preventing NLRP3 inflammasome activation.
Previous Findings from the COVID-OUT Study COVID-OUT primarily aimed to evaluate the efficacy of metformin, ivermectin, and fluvoxamine in preventing severe disease and reducing death risk in outpatient COVID-19 patients. The study included patients who had been diagnosed within three days and had experienced symptoms for no more than seven days. Participants were aged 30 to 85 and were either overweight or obese.
The primary endpoint was a composite of hypoxemia (oxygen saturation ≤93% measured by a home pulse oximeter), emergency visits, hospitalization, or death. Initial results, published in NEJM in August 2022, showed no significant differences between metformin, ivermectin, fluvoxamine, and placebo. The odds ratios (OR) for the primary endpoint were 0.84 (P=0.19) for metformin, 1.05 (P=0.78) for ivermectin, and 0.94 (P=0.75) for fluvoxamine.
However, due to the ambiguity of using a single oxygen saturation measure >94% to define severe COVID-19, the study included a secondary composite endpoint, excluding hypoxemia. Results showed that metformin reduced the secondary endpoint risk by 42% (OR 0.58), while ivermectin (OR 1.39) and fluvoxamine (OR 1.17) did not show improvement. When focusing only on hospitalization and death risk, metformin reduced the risk by 53% (OR 0.47), ivermectin by 27% (OR 0.73), and fluvoxamine showed no benefit (OR 1.11).
In a Lancet Infectious Diseases article published in 2023, the researchers analyzed the incidence of long COVID after 10 months of follow-up. The long COVID rates for metformin and placebo groups were 6.3% and 10.4%, respectively, with metformin significantly reducing the risk by 41% (HR 0.59, P=0.012). In patients who took metformin within three days of symptom onset, the risk of long COVID was reduced by 63% (HR 0.37). Ivermectin (HR 0.99) and fluvoxamine (HR 1.36) did not show preventive effects against long COVID.
How Effective is Metformin as an Antiviral? What About the Impact of Vaccination? The authors also compared metformin’s antiviral effects with other small-molecule drugs like Nirmatrelvir and Molnupiravir. These non-head-to-head comparisons revealed different antiviral effects at various time points for each drug.
Does Vaccination Affect Antiviral Efficacy? Vaccination can influence the efficacy of antiviral drugs. In the COVID-OUT trial, nearly half of the patients were vaccinated. Among vaccinated patients, metformin’s antiviral effect on day 10 (-0.48 log10 copies/mL) was lower than in unvaccinated individuals (-0.86 log10 copies/mL). This likely reflects the presence of memory B and T cells in vaccinated individuals, which may respond quickly early in infection, reducing the need for antiviral drugs. Nonetheless, metformin remained effective in preventing severe outcomes like emergency visits, hospitalization, or death in vaccinated individuals, suggesting it could offer additional protection even in this population.
Does the Dosage of Metformin Impact Antiviral Effects? The study observed that metformin’s antiviral effects increased with dosage. On day 5, at a dose of 1000 mg per day, the antiviral effect was -0.47 log10 copies/mL. By day 10, when the dose increased to 1500 mg daily, the effect improved to -0.64 log10 copies/mL. In clinical practice, metformin doses are titrated slowly, usually reaching 2000 mg per day over a period of 4 to 8 weeks for chronic conditions like diabetes and obesity.
