Skip to Main Content

Pfizer released news late Friday that Paxlovid, the antiviral currently subject to a big push from the U.S. government, failed to prevent people living with Covid patients from catching the infection.

The news is one of several bad headlines for the new Covid pill, but one experts say doesn’t affect the medicine’s primary use: treating people who are already sick.


Paul Sax, clinical director of the division of infectious diseases at Brigham and Women’s Hospital, said he would “absolutely” prescribe Paxlovid to people at high risk of severe disease who have Covid. “Without hesitation,” he said. “Because the net benefit in the high risk study was extremely high.”

And some experts said the results, while disappointing, are not a surprise. “Traditionally it’s been difficult to use small molecule antivirals for true prophylaxis because the biology of treating infection is different from the biology of preventing infection,” said Daniel Barouch, director of the Center for Virology and Vaccine Research at Beth Israel Deaconess Medical Center.

In the new study, Paxlovid or placebo was given to 2,957 adults who had a negative rapid antigen test and were asymptomatic, but who were living with someone who had Covid. The medicine (or the placebo) was given within 96 hours of the infected individual becoming infected. Volunteers were randomly assigned to receive either a five-day course of Paxlovid, a 10-day course, or placebo.


Patients who received the five-day course tested positive 32% less often for SARS-CoV-2, the virus that causes Covid, than those who received placebo. Those who received the 10-day course tested positive 37% less often than those who received placebo. But both of those results were not statistically significant, and may have been due to chance.

The results contrast sharply with those for monoclonal antibody drugs earlier in the pandemic, where roughly similar studies showed Regeneron’s REGEN-CoV, Eli Lilly’s bamlanivimab, and AstraZeneca’s Evusheld all reduced rates of infection by 80%.

Figuring out why may be difficult, especially until more data are available.

“I would be very hesitant to draw a lot of conclusions from a press release,” said Myron Cohen, director of the Institute for Global Health & Infectious Diseases at the University of North Carolina Medical School. Cohen helped conduct the Regeneron and Lilly studies.

Eli Lilly conducted a prophylaxis study of its monoclonal antibody in nursing homes before vaccines were available. Receiving the antibody reduced the risk of contracting Covid by 80%. Regeneron’s study was in household contacts, and separated those who tested negative via PCR and for SARS-CoV-2 antibodies from those who were positive. The first group were protected 80%; the second had milder symptoms than they otherwise would. But both of those antibodies are not effective against newer variants. The Evusheld study was the most recent, but still occurred before the Omicron wave.

One possibility, Cohen said, is that the study failed the drug: that is, Paxlovid would have been effective if it had been given earlier, or that people who tested negative on antigen tests were already infected with the virus but did not yet have symptoms.

It’s also possible that Paxlovid didn’t perform differently than a monoclonal antibody would, said Barouch, who also worked on the Regeneron and Lilly studies. The studies of prophylaxis with monoclonals were done before the Omicron strain was prevalent, and Omicron is more infectious. “That hyper-infectiousness probably makes it more difficult to prevent infections,” said Barouch.

But it also could have to do with differences between the way Paxlovid and the monoclonals work. It could be, Cohen said, that the drug doesn’t get to the mucosa of the nose, where infection occurs, fast enough.

Or it could be because monoclonals work by preventing the virus from entering cells, whereas Paxlovid works by inhibiting the virus’ replication, blocking an enzyme known as a protease.

“I think it may have to do with the mechanism of action of the drug, which acts relatively late in viral replication,” said Sax. “So perhaps once the virus has established infection, blocking the protease enzyme is too late to prevent infection from taking hold.”

Barouch echoed this idea. He also connected it to another hot issue with the drug when it is used for treatment, not prevention: anecdotal reports that some people have initial success being treated with Paxlovid but then see their virus levels rise and symptoms return. It is as if, he said, the drug is not eradicating the virus, but just suppressing it.

He emphasized that this would not give him any pause in treating a positive patient who has risk factors for severe Covid.

“If someone who is moderate-to-high risk acquired SARS-CoV-2,” said Barouch, “these new data would not make me less enthusiastic to prescribe Paxlovid for therapeutic purposes.”

Sax has written about the relapses after Paxlovid treatment. Most people who are given the drug, he emphasized, are not relapsing, and that means they are getting a benefit with regard to their own health and may also be less likely to spread the virus to other people. But he said more data are needed on how often relapses occur with treatment. It’s not clear, he said, whether they should get a longer treatment course, something that might need to be tested in a clinical trial, or, in some cases, a second course of treatment. For mild recurrences, though, retreatment may not be justified, because people get better.

Sax emphasized that these concerns don’t change his willingness to use the drug when it has been shown to lower hospitalization and death. Most anti-infective drugs won’t show that in a clinical trial of healthy people, he noted, simply because people generally do get better. Another benefit from using the medicine could simply be to help people recover more quickly.

Cohen said that he is more concerned about people who, perhaps because they are immunocompromised, might not be able recover entirely even with Paxlovid. He said there is a need to try the drug in combination with monoclonal antibodies, with Merck’s molnupiravir, or with other potential treatments in those patients.

So far, clinical trials of Paxlovid have not included vaccinated patients, but all three doctors said that they were comfortable prescribing the drug to people who had risk factors for severe disease, such as age, diabetes, or obesity, who had been fully vaccinated.

Create a display name to comment

This name will appear with your comment

There was an error saving your display name. Please check and try again.