An anti-malaria drug that has been trumpeted as a therapy for COVID-19 was unable in a University of Minnesota clinical trial to prevent the onset of the infectious disease.

The results of the nation's first randomized trial with the drug, hydroxychloroquine, against COVID-19 will disappoint doctors who had hoped for new therapies amid the pandemic. Many prescribers had given it off-label to COVID-19 patients — in the absence of other options — and President Donald Trump had been an early champion of the drug and said in mid-May that he was taking it for the preventive benefit, a benefit that the U study could not verify.

"While we are disappointed that this did not prevent COVID-19, we are pleased that we were able to provide a conclusive answer," said Dr. David Boulware, leader of the U trial. "Our objective was to find an answer."

Results published online Wednesday by the New England Journal of Medicine showed little difference in the onset of COVID-19 in 414 people who took hydroxychloroquine and a comparison group of 407 that took only folic acid vitamins. All participants had at least moderate risk for COVID-19 after being exposed to others in their homes or workplaces who had the illness.

There was a small difference, as 11.8% of people taking the drug developed COVID-19, compared with 14.3% of those taking vitamins, the study showed. However, that difference was considered statistically insignificant.

Even if valid, the small difference means that 42 people would have to take the drug prophylactically to prevent one COVID-19 case. That would be costly and expose people to drug side effects. The study showed that 40% experienced mild side effects such as nausea, but it allayed fears that widespread use of the drug could result in cardiac and other complications.

The study provides evidence only of whether the drug is effective in preventing COVID-19 onset following exposure to someone with the infection.

A separate U trial is examining whether it benefits patients after early onset of symptoms, but results haven't been published.

Enthusiasm for the drug has already cooled after open-label trials and case reports that found little benefit, but the U results are a significant advance.

Minnesota as of Wednesday reported 25,870 lab-confirmed cases and 1,086 deaths from COVID-19, which is caused by a novel coronavirus for which there is no vaccine.

While COVID-19 case numbers have ebbed, state health officials suspect another wave after protests over the death of George Floyd brought thousands of people together.

Chanting and shouting could hasten the spread of the virus, said Jan Malcolm, state health commissioner, along with police use of tear gas to break up protests that could cause infected people to breathe heavily.

"COVID-19 can spread easily and quickly in large groups of people who are close together for long periods of time," she said.

The state is advising protesters, first responders and community volunteers to seek diagnostic testing for COVID-19, ideally from their own doctors, five to seven days after participation in mass events.

Anxiety about COVID-19 was reflected last week in a study in the Journal of the American Medical Association, which showed a twentyfold surge in prescriptions of hydroxychloroquine and a similar drug, chloroquine, in late March after the World Health Organization declared a COVID-19 pandemic and the president tweeted March 19 in favor of the drug.

Surges based on hype threaten the supply of the drug for approved treatments of rheumatoid arthritis and lupus, said Dr. Frank Rhame, a virologist at Allina Health, which is why the U study results are so important.

"It's a great illustration of the importance of getting a well-designed study to get at the medical reality of the situation," Rhame said. "You have to be open to all possibilities until you have the data."

The U trial was the first in the world to assess whether hydroxychloroquine could prevent the onset of COVID-19 in people exposed to the novel coronavirus that causes it, and the first placebo-controlled double-blind study of the drug's effect on COVID-19. Blinding means that patients didn't know whether they took vitamins or drugs, and researchers didn't know either.

While not a primary goal, the study also found no increased protection for participants who also took zinc or vitamin C. A separate study suggested that zinc could boost the effectiveness of chloroquine, but even if that's true, the U authors noted that most Americans consume enough zinc already so supplements shouldn't be needed.

Jen Reiner of Edina figures she took the real hydroxychloroquine during the study, based on the continuous nausea she felt. She didn't get COVID-19 despite being in close contact with her infected husband, but then neither did her children and they weren't in the study.

"A bit bummed the drug didn't seem to work," Reiner said, "but it had to be one way or the other."

Boulware said the results were validated and held up to extensive peer review but may face challenges and criticisms from skeptics hoping for better results. The researchers acknowledged limitations of the study, which recruited participants in North America online and mailed medication to them. A global shortage of diagnostic testing supplies at the time meant that researchers had to count some cases of COVID-19 in the study group based on symptoms alone.

"I'm sure people will come after it," Boulware said, "but the data are what the data are."

Critics on one side felt hydroxychloroquine was a wonder drug and that all study participants should receive it, while critics on the other side felt it was too dangerous to study, Boulware said. "In reality, it's neither of them."

The U also participated in a national trial of remdesivir, which has helped COVID-19 patients and received emergency use authorization by the U.S. Food and Drug Administration. Supplies of the drug have been distributed by the federal government, which to date has sent 5,840 vials to Minnesota — enough to treat 500 to 1,000 people.