COVID-19 patients were less likely to die or need hospital care if they took the common anti-diabetes drug metformin, according to a long-awaited University of Minnesota study, but not if they took ivermectin.

The results of the ambitious remote clinical trial could steer doctors toward off-label use of one cheap and available drug for the treatment of COVID-19 — just not the one that has drawn the most public attention. Results were presented in a national online lecture on Friday ahead of publication in a medical journal.

"It does appear that metformin substantially reduces the risk of [emergency department] visits, hospitalization or death from COVID-19, and that reduction is significant," said Dr. Carolyn Bramante, the leader of the U's COVID-OUT trial, in an interview. The study also examined whether the antidepressant fluvoxamine offered any benefit.

The study of more than 1,300 participants is a third strike against ivermectin, a controversial anti-parasitic drug that also failed to produce a benefit against COVID-19 in large clinical trials in the U.S. and Brazil. However, the finding in the U trial was not statistically significant, meaning that it neither proved that ivermectin worked or didn't work when taken over three days in a moderate range of dosages.

"All I can say is that, in our study, we did not see any evidence of benefit of ivermectin in those doses," Bramante said.

Most of the results failed to reach statistical significance in the yearlong trial, which limited enrollment to people who were overweight in order to study the drugs in a known high-risk group for severe COVID-19. The primary goal was to see if patients had lower composite scores after taking the three drugs, alone or in combination, that were based on whether they died, needed hospital care, and suffered abnormally low blood-oxygen levels.

Practical problems got in the way of the composite results, including the use of commercial blood-oxygen monitors that turned out to have unreliable readings. Findings were more clear cut on whether the drugs prevented COVID-19 deaths, hospitalizations or ER visits.

Low doses of fluvoxamine showed such little evidence of benefit that the arm of the trial involving that drug was stopped early. Patients taking higher doses of ivermectin were more likely to need hospital care for COVID-19 than those taking lower doses, but results for both were within the statistical possibility of random chance.

All three drugs were selected for the study based on findings early in the pandemic that they could inhibit replication of the coronavirus or the inflammation that causes COVID-19 disease. All three also offered promise as low-cost existing medications with established safety records.

Ivermectin has gained a fervent following based in part on observational clinical successes against COVID-19 reported by a Florida clinical group and its promotion as an alternative by people with vaccine safety concerns.

Metformin was selected through a U analysis that identified cellular targets of the coronavirus and then found drugs that worked on those same targets. The drug's performance actually improved over time as the pandemic shifted from an alpha coronavirus variant last spring to faster-spreading delta and omicron coronavirus variants this winter.

"That is reassuring. It appears metformin doesn't lose relevance," Bramante said, although results among the different variants didn't reach a level of statistical significance.

Pandemic trends uncertain

Minnesota's pandemic indicators are mixed regarding next stages of COVID-19. The viral load found in sewage samples declined 11% over the past week at the Metropolitan Wastewater Treatment Plant in St. Paul — after rising slightly the previous two weeks.

The Centers for Disease Control and Prevention this week also found zero Minnesota counties with high enough COVID-19 levels to recommend public mask-wearing indoors. The entire Twin Cities area dropped from moderate to low risk levels as well.

On the other hand, the wastewater data showed that most COVID-19 in the Twin Cities is now caused by a BA.5 coronavirus subvariant. While the variant has caused lower rates of severe illness than others, it also shows a high breakthrough rate in people with immunity from recent infections and vaccinations. Mayo Clinic's 14-day forecast suggests infection levels will rise as a result.

"There is the potential that it could become a larger more sustained increase," said Curtis Storlie, a co-creator of the Mayo COVID-19 model, "but it is also possible that this current rise loses steam in the next couple of weeks and we wait to catch another 'wave' in the coming months."

Health officials are confident that broad immunity levels will continue to reduce the rate of coronavirus infections that produce severe COVID-19. Minnesota also has a growing stockpile of antiviral drugs that didn't exist last year. The state this week reported an on-hand supply of more than 18,000 courses of Paxlovid, the most effective COVID-19 antiviral drug.

The U's COVID-OUT trial was selected for the national presentation Friday because of its unique design as a national remote clinical trial studying multiple drugs in combination. Bramante said the research team went to great lengths to immediately ship study drugs to participants with confirmed cases of COVID-19 so they could start taking them as soon as possible.

Patients on average received their study drugs within 24 hours of enrollment, but still weren't taking them until five days after symptom onset, which could have hindered results.

"Our time to study drug initiation was a little longer than it could be," Bramante said, "if we were trying to mimic real life as much as possible."