Editor’s note: Find the latest COVID-19 news and guidance in Medscape’s Coronavirus Resource Center.
Treatment with the diabetes drug metformin shows a significant, dose-dependent effect in lowering SARS-CoV-2 viral load within days of administration, according to the latest analysis of the phase 3 COVID-OUT trial. These findings add to a multitude of benefits the drug has been shown to have in COVID infection.
As previously reported by Medscape Medical News, COVID-OUT did not meet its primary endpoint, but it did show important secondary outcomes including a 42% reduction in emergency room visits and in hospitalizations and/or deaths by day 14, and a 58% reduction in hospitalizations/death by day 28. A further subanalysis has shown a 42% reduction in long COVID compared with placebo.
“In this phase 3 randomized controlled trial, metformin showed prevention of severe COVID, prevention of Long COVID, and an antiviral effect, and this is consistent with other data,” said coauthor Carolyn Bramante, MD, University of Minnesota, Minneapolis, in presenting the findings at the Conference on Retroviruses and Opportunistic Infections (CROI) 2023 Annual Meeting.
For the new subanalysis, the authors further evaluated the effects of metformin treatment on SARS-CoV-2 viral load.
A total of 1323 patients in the study, enrolled at six centers, were randomized to treatment either with metformin 1000 mg per day on days 2 to 5 and 1500 mg per day on days 6 to 14 (n = 187), or to ivermectin 390-470 mcg/kg per day for 3 days (n = 187), fluvoxamine 50 mg twice daily for 14 days, and/or an exact-matching placebo in a 2 x 3 factorial trial design.
The subanalysis on viral load included 483 patients from the trial who were treated with metformin versus 462 who received placebo, who were all enrolled within 3 days of a documented SARS-CoV-2 infection and less than 7 days after symptom onset.
The patients had a median age of 46 years, and all had either overweight or obesity. Only about 2% had diabetes, and only patients considered low-risk were excluded from the trial, including those under age 30 and those with a body mass index under 25.
About half of patients had received a primary vaccine and about 5% had received a vaccine booster. SARS-CoV-2 variants that were prominent during the study included Alpha, Delta, and Omicron.
The viral samples available on days 1, 5, and 10 showed a mean change in viral load from baseline to follow-up; the viral load was significantly lower with metformin versus placebo (-0.64 log10 copies/mL), representing a 4.4-fold greater decrease in viral load with metformin.
The mean rate of undetectable SARS-CoV-2 viral load at day 5 was 49.9% in the metformin group versus 54.6% in the placebo group (odds ratio [OR], 1.235), and the undetectable rate at day 10 was 14.3% in the metformin group and 22.6% in the placebo group (OR, 1.663; P = .003).
An increased antiviral effect corresponded with increases in metformin dosing on days 6 through 14. Furthermore, the antiviral effect became stronger when metformin was started earlier in the course of infection.
Of note, the antiviral effect was more pronounced among those who were not vaccinated (mean -0.95 log copies/mL) compared with the vaccinated (mean -0.39 log copies/mL).
The antiviral effect with metformin was similar to that seen with nirmatrelvir at day 5 and was greater than nirmatrelvir at day 10.
No similar relationships in SARS-CoV-2 viral load were observed between ivermectin or fluvoxamine and placebo.
The findings are consistent with results of other recent observational studies, including research showing metformin to be associated with reductions in COVID-19 severity in patients with prediabetes, Bramante noted.
The authors’ previous analysis looking at long COVID in the COVID-OUT study showed that metformin treatment during acute COVID significantly reduced the risk for a diagnosis of long COVID versus placebo at 300 days following randomization, with a hazard ratio of 0.59 after adjustment for the study drug and vaccination at baseline.
Bramante noted that metformin’s potential antiviral properties have long been speculated, with some of the earliest research on the drug suggesting less severe outcomes in influenza, and more recently, RNA assays suggesting effects against other RNA viruses, including the Zika virus.
In terms of COVID, Bramante noted that the drug has plenty of potentially favorable benefits.
“Metformin is very safe and is known to have very few contraindications, so the next steps could be to consider looking at this in terms of a combination therapy,” she said.
“Data From Other Studies Are Conflicting”
Commenting on the study, Diane V. Havlir, MD, cautioned that “metformin is currently not recommended in treatment guidelines, [and] data from other studies are conflicting; side effects can be an issue, and the study presented here was in a select population,” she told Medscape Medical News.
However, “what is both new and interesting in this presentation is the reduction of viral load, which [was observed] in the samples collected not only on days 1-5, but also days 6-14,” said Havlir, who is professor and associate chair of Clinical Research, Department of Medicine, chief of the Division of HIV, Infectious Diseases and Global Medicine, and director of the AIDS Research Institute at the University of California San Francisco.
Key questions the findings raise include whether the results correlate with clinical outcomes or transmission, and whether the findings are generalizable to other populations and settings, Havlir said.
Ultimately, “we need to continue to pursue all aspects of outpatient treatments for COVID to address questions like these for new and existing agents,” she added.
Conference on Retroviruses and Opportunistic Infections (CROI) 2023 Annual Meeting. Abstract #170. Presented February 21, 2023.
The trial received funding from the Parsemus Foundation, the Rainwater Charitable Foundation, Fast Grants, and the United Health Group. The authors and Havlir have disclosed no relevant financial relationships.
For more news, follow Medscape on Facebook, Twitter, Instagram, YouTube, and LinkedIn.