To Avoid Paxlovid Rebound, Some Experts Call for Longer Courses of Treatment

Prescriptions for the antiviral drug Paxlovid have soared since the medication was granted emergency use authorization last December. High-profile people who recently tested positive—including President Joe Biden, White House medical advisor Dr. Anthony Fauci, and late night host Stephen Colbert—have also taken the drug.

But some people—including those three famous patients—have reported rebound infections after taking the pills. Paxlovid rebound occurs when a person takes the drug for a few days, tests negative, and then tests positive again several days later.

The growing number of cases is prompting questions about whether people should be taking Paxlovid for longer periods of time to avoid rebound infections. Here’s what virologists and the latest studies say.

How common is Paxlovid rebound?

Researchers have established that people who test positive after finishing a five-day course of Paxlovid are not contracting a new infection. Genetic studies show that the virus that comes back soon after a person stops treatment is the same one that caused the initial infection. But it’s not yet clear how frequent rebound infections are. In the original studies submitted to the U.S. Food and Drug Administration (FDA) for emergency use authorization, Paxlovid’s drugmaker Pfizer found that rebounds happened in 1-2% of patients—the same rate as in the placebo group.

Dr. David Ho, professor of medicine and director of the Aaron Diamond AIDS Research Center at Columbia University, is studying Paxlovid rebound and believes the prevalence is likely higher. “In my own experience, I have now counted 15 friends, family members, and colleagues who have taken Paxlovid, and over half have rebounded,” he says. Though that’s not a scientific tally, “physicians with large COVID-19 practices will tell you that it’s not rare.”

Figuring out exactly how often rebound occurs may not happen any time soon. If rebound doesn’t frequently occur, tens of thousands of people taking the drug would have to be followed in order to adequately determine how often people test positive again. “I’m afraid that will never be done,” says Dr. Mark Siedner, a clinical epidemiologist at Massachusetts General Hospital and associate professor of medicine at Harvard Medical School, who has studied patients with Paxlovid rebound.

To get at least some idea of the frequency, Siedner’s group is enrolling a few hundred people in a study that will hopefully shed some light on the question. The volunteers will get their noses swabbed three times a week and the researchers will analyze the samples to look for any live virus, and if viruses are present, grow them to see if they continue to remain active and infectious. His team has already done a small study showing that people with Paxlovid rebound remain contagious when they test positive the second time, which means they need to extend their isolation period for another five days at least, and continue wearing masks during that time if they come into contact with other people.

Absent large and conclusive data, Siedner says health officials at the FDA would have to weigh the benefits and risks of Paxlovid in considering whether doctors should be able to prescribe a longer course of treatment.

The case for longer treatment

Is there justification for a longer course of treatment? “In my mind, there is,” says Ho—who, in addition to studying Paxlovid rebound, has experienced it himself. In studies, Ho says he’s observed that Paxlovid does its job of suppressing SARS-CoV-2, halting the virus from continuing to replicate and essentially trapping it in an intermediary form. This version of the virus isn’t fully formed, and needs a few more steps of development before it becomes active and can infect new cells. Paxlovid suppresses the virus at this stage. But in Ho’s studies, the half life of the virus in this suspended state is nearly a day—anywhere from 19 to 22 hours—which means it takes that long for about half of the virus to decay. For someone who is infected, if there is still enough of this intermediary form in the body after the fifth and final day of Paxlovid treatment, that virus could reactivate, finish its development, and start infecting cells anew—thus causing a rebound infection.

“The virus is rather persistent,” says Ho. “And we believe that five days of treatment is not enough to have that form decay so that it’s nonexistent at the end of those five days.”

Based on his lab results, Ho’s team calculated that extending Paxlovid for several more days—three to five days beyond the current regimen—could reduce the risk of rebound by 10-fold. The additional few days would eliminate more of the intermediary form of the virus circulating in the body. More studies are needed to validate that, says Ho, and to confirm that extending the drug for several more days is safe.

Paxlovid isn’t the only antiviral drug that acts this way. Paxlovid is a protease inhibitor—meaning it interferes with the virus’ ability to form the final proteins it needs—and other medications that belong to this class of drugs cause similar infection rebounds. For example, says Ho, ensitrelvir, a SARS-CoV-2 antiviral that the Japanese pharmaceutical company Shionogi is studying, produced similar intermediaries that hadn’t deteriorated before the drug was stopped and bounced back to cause infection again. “It’s related to the mechanism of action of the drug, not related to a deficit in [Paxlovid],” says Ho.

It’s similar to the way antibiotics work: if people stop taking antibiotics before the prescribed time period is up, the infection can return before the enough of the bacteria is eliminated. “We’ve certainly seen infections where if you give the treatment, you knock down the level of bacteria or virus, but if you don’t completely eradicate it—and then it comes roaring back,” Siedner says. With Paxlovid rebound, “my guess is that it’s happening because people aren’t getting treated long enough.”

Some people infected with SARS-CoV-2 who haven’t taken Paxlovid have even experienced rebound infections, in which they test negative and think they’ve recovered, but then test positive several days later, Siedner says. But the situations are different. “People with Paxlovid rebound have much higher viral loads, the viral load stays high for a much longer period of time, and the symptoms last longer,” compared to people who rebound and haven’t taken the drug, he says. “Paxlovid rebound is the real deal.”

In a statement, a Pfizer spokesperson said that because the company’s initial studies found similar rebound rates in the treatment and placebo groups, those data “suggest the return of elevated detected nasal viral RNA (also known as viral rebound or COVID-19 rebound) is uncommon and not uniquely associated with any specific treatment.”

Ho notes, however, that those studies were not specifically designed to detect and measure rebound, and that more detailed analyses are needed.

Besides taking a longer course of Paxlovid, another option for addressing rebound would be to allow patients to start a second course of Paxlovid once they test positive again, experts say. The Pfizer spokesperson confirmed that the company is working with the FDA to set up a trial studying this.

The case against a longer course of treatment

Like any drug, Paxlovid has side effects—the most common of which is an extremely bad taste in the mouth that temporarily affects some people’s ability to eat. The medication also interacts with a number of other frequently used drugs, including cholesterol-lowering statins. “We often ask people to hold off on some of their other medications or decrease doses of their medications for five days while taking Paxlovid,” says Dr. Davey Smith, professor of medicine at University of California, San Diego, who has studied Paxlovid rebound in patients. “Asking them to hold off longer may have more risks and outweigh the benefits of doing that. There are a whole bunch of complications in just asking somebody to take the drug for 7 or 10 days.”

Siedner also notes that when looking at the benefits versus the risks of longer therapy, doctors need to consider how much people who rebound are impacted by the second infection. In a study from Kaiser Permanente of more than 5,000 people who took Paxlovid, fewer than 1% ended up in the hospital because of their symptoms—including those who rebounded. “By and large, the health record systems are not seeing huge numbers of people needing hospitalization that could be avoided with longer treatment,” he says. “If rebound isn’t happening that often, and people who rebound aren’t at higher risk of hospitalization than people who had not taken Paxlovid, then I think there may not be much benefit to giving everyone a longer course of treatment.”

Another fear of extending the course of Paxlovid is that SARS-CoV-2 could potentially become resistant to the drug. Ho reported the first evidence of this in a paper published Aug. 8 in pre-print form on the bioRxiv server. In lab studies, he and his team found that when the virus was challenged with doses of Paxlovid, it developed multiple pathways to evading the drug, raising concerns about the need to design next-generation antivirals or to use them in combination, as doctors currently do to control HIV. More studies are needed to document how quickly the virus develops resistance to Paxlovid and whether extending the therapy by a few days would increase this risk.

Is it still worth taking Paxlovid?

For people who have risk factors that make them more vulnerable to COVID-19, doctors continue to say that the benefits of Paxlovid outweigh the risk of rebound, since the drug protects people from getting sick enough to need hospitalization and from dying of COVID-19. Most people experiencing rebound report relatively mild symptoms and don’t require more intensive medical care. Even if some people won’t be able to suppress their infection completely after a course of Paxlovid, the treatment may still be helping them to avoid more serious disease. As more data become available, health officials may reevaluate how long that round of treatment needs to be.