Why Cheap, Older Drugs That Might Treat Covid Never Get Out of the Lab
In March 2020, Dr. Joseph Vinetz left the contemplative world of his Yale University infectious-disease laboratory and plunged into the covid ward at Yale New Haven Hospital, joining an army of health care workers who struggled to treat the deadly viral disease.
There were no drugs against covid-19, and no way to predict which infected patients would develop pneumonia or fall into an inflammatory tailspin leading to severe illness or death. In desperation, Vinetz and countless other doctor-scientists trawled the literature for existing medicines that might help.
“We were in the hospital. We had nothing,” Vinetz said. “I was one of tens of thousands of doctors around the world who said, ‘We gotta figure out what to do.’”
On April 16, 2020, Vinetz saw an article in the journal Cell about a drug called camostat, licensed in Japan in 1985 to treat inflammation of the pancreas. Research during the first SARS epidemic, in 2004, had shown the drug had a plausible biochemical mechanism for slowing coronavirus infections, so Vinetz and his colleagues quickly organized a small clinical trial on outpatients with mild to moderate symptoms.
In those days, before covid vaccines and covid-specific treatments appeared on the market, Vinetz’s experiment was one of thousands conducted by doctors who hoped older vaccines and drugs, usually cheap and off-patent, might provide them with options.
Mostly, the drugs were too toxic or had no clear effect. Of the more than 1,500 trials for potential covid drugs listed on the website of the National Institutes of Health — including antivirals, anti-inflammatories, and drugs used for cancer, asthma, heart disease, and dozens of other conditions — few have produced helpful medicines.
In fact, only one older drug is routinely used to fight covid. That’s the steroid dexamethasone, proven by British scientists to help keep hospitalized patients from requiring supplemental oxygen or intubation.
Drugs like hydroxychloroquine and ivermectin showed hints of value initially but failed in clinical trials — only to remain in circulation, at least partly because their use symbolized affinity in the culture war for some of President Donald Trump’s followers.
A few old drugs still show promise, but they’ve had trouble getting traction. The ivermectin and hydroxychloroquine fiascoes soured doctors on repurposed medications, and the pharmaceutical industry has shown little interest in testing them, especially when it can earn billions from even mediocre new ones, scientists tracking the field say.
American and European scientists have confirmed the theoretical basis for camostat’s impact on covid. But evidence for its effects is weak; last year the drug was dropped from a big NIH trial comparing various treatments.
A more promising story emerged with fluvoxamine, licensed under the brand name Luvox in 1994 to treat obsessive-compulsive disorder. The drug is in the same class as common antidepressants such as Prozac, Lexapro, and Zoloft.
A child psychiatrist noticed fluvoxamine might be good for covid. In March 2020, while recovering from a bout of covid, Dr. Angela Reiersen of Washington University in St. Louis saw a 2019 study in mice that showed how fluvoxamine could activate a protein similar to one missing in patients with Wolfram syndrome, a genetic disease that causes diabetes, neurological issues, and, eventually, death.
Reiersen and her colleague Dr. Eric Lenze, a geriatric psychiatrist, began a clinical trial of the drug in people with symptoms of covid. Of the 80 in the fluvoxamine group, none suffered a serious decline, while six of 72 patients given sugar pills got pneumonia, and four were hospitalized.
In a follow-up 1,500-patient trial in Brazil, people who took at least 80% of their fluvoxamine pills were 66% less likely to require emergency care or hospitalization than those who got sugar pills. Only one died, compared with 11 in the placebo group.
Since October, when the Brazilian study was published, fluvoxamine’s future has dimmed. Neither the NIH nor the Infectious Diseases Society of America recommends fluvoxamine to prevent respiratory distress. The NIH panelists noted that the better outcomes in the Brazilian trial were only statistically significant among those who remained in the trial. (Because of nausea and other side effects, only 74% of trial participants in the fluvoxamine wing took all their pills, compared with 82% in the placebo wing.)
The NIH panel also was put off by the fact that the Brazilian trial counted hospitalizations as well as people put under a doctor’s care for six hours or more — not a standard measure. Trial organizers said that was necessary because Brazilian hospitals were so packed with covid patients that many people got their care in makeshift outdoor shelters.
Regulators and experts are awaiting results from two other big trials, one organized by a consortium of universities and hospitals, the other by the NIH. But both studies are using doses of 100 milligrams of fluvoxamine a day, compared with 200 or 300 milligrams in the successful trials.
“I have concerns that they are not using a high-enough dose,” Reiersen said, given that fluvoxamine operates on a different biochemical pathway to fight covid than the one involved in psychiatric treatment.
The concern is shared by Craig Rayner, a former drug company scientist who worked on the Brazilian trial and other big tests of repurposed drugs. “You can do the largest, most well-funded study in the world,” he said, “but if you choose the wrong dose, it’s rubbish in, rubbish out.”
The team overseeing NIH’s trial opted for a lower dose because higher doses had already been used in the earlier trials — and often caused side effects, said Sarah Dunsmore, a program director at NIH’s National Center for Advancing Translational Sciences.
On Dec. 21, David Boulware, a University of Minnesota infectious-disease expert, petitioned the FDA to approve a change in fluvoxamine’s label stating it can be used to prevent respiratory distress in at-risk patients with mild to moderate covid. He hasn’t received a response yet.
It’s a different story for big drug companies. Two days after Boulware’s submission, FDA authorized Merck to market its drug molnupiravir, which in its clinical trial showed about as much effectiveness as fluvoxamine, and also had side effects like nausea and dizziness. Fluvoxamine also can cause insomnia and anxiety; molnupiravir is not recommended for pregnant women or anyone, male or female, having unprotected sex, because it caused genetic and fetal damage in test animals.
Still, federal guidelines recommend molnupiravir in certain settings, and the government has bought more than 3 million doses for about $2.2 billion, or $733 per dose. Fluvoxamine, a generic, goes for less than $5 a pill.
“You hate to say that Big Pharma has a lot of influence, but clearly they do,” Boulware said. “The molnupiravir data was not that great, but we’re spending billions on the drug and it got fast-track emergency use authorization” while fluvoxamine remains in a gray area.
With the arrival of effective vaccines and the trickle of antiviral treatments, the urgency of rehabilitating old drugs for U.S. patients has ebbed. But the need remains high in lower- and middle-income countries where vaccines and new covid treatments remain unavailable.
It’s not rare for a pharmaceutical company to synthesize or study a drug for one purpose, only to discover it works better for something else. The classic instance is sildenafil, or Viagra, which was being developed as a drug for hypertension when scientists noticed a remarkable side effect. Remdesivir, now a front-line drug against covid, was aimed at treating Ebola.
It’s less common for a drug marketed for one use to acquire an entirely different purpose, but the pandemic drove scientists to try. They tested thousands of compounds in petri dishes for their virus-killing power, but the journey from test tube to human remedy is long, said Rayner, who is also a professor of pharmaceutical sciences at Monash University in Melbourne, Australia.
If fluvoxamine were a new drug, the company sponsoring it would have spent the money needed to get the drug approved and to show the FDA it has the means to monitor the drug’s safety and efficacy. Since it’s an old drug, it will be up to independent scientists, or perhaps a reluctant generics manufacturer, to sponsor safety monitoring should the FDA provide an emergency use authorization, Rayner said.
An EUA or approval “comes with strings. You have to continue to monitor the safety, to make sure no signals pop up when you move it from thousands to millions of patients,” he said. “That’s very expensive.”
U.S. physicians can prescribe drugs off label, but most are leery of doing so until a drug has won approval for the new use. That’s especially true now.
Definitive answers on some repurposed drugs were slow in coming because there were too many small, poorly designed studies by “every man and his dog,” Rayner said. He calculates up to $5.6 billion has been wasted on hydroxychloroquine clinical trials alone.
A recent World Health Organization resolution called for better coordination and information-sharing among those organizing trials so that definitive answers can be obtained quickly with big pots of data.
As for camostat, Vinetz said those who took the drug felt better than those who got a placebo. “It basically prevented loss of smell and taste, which people really bitterly care about,” he said. “That means there’s a real biological effect. That merits further exploration.”
But will that happen? Vinetz’s team has sought publication of their research for five months with no success. He’d like to see whether camostat can prevent long covid, but such investigations cost millions. Camostat’s Japanese manufacturer apparently lost interest in it as a covid drug after its own small, unsuccessful trial.
“When there’s no profit motive, it’s tough,” Vinetz said. Meanwhile, he’s resumed his research into controlling a neglected tropical disease: leptospirosis.