Paxlovid, Pfizer’s antiviral treatment for COVID-19, is turning out to be one of the most successful new drugs in pharma history. Prescriptions have increased tenfold since February, and the five-day pill pack accounts for nearly 90% of the COVID-19 oral-antiviral market.

Based on committed contracts from governments around the world, sales of the drug will reach $22 billion this year, Pfizer affirmed May 3.

President Joe Biden’s administration has made Paxlovid a pillar in its plan to return the U.S. to normal life. The government’s test-to-treat program is intended to put the pills into the hands of every American who tests positive for COVID-19.

Yet somehow, there seems to be no unified plan for carefully tracking who exactly benefits from the drug and who doesn’t. Such monitoring is urgently needed.

Paxlovid is extraordinarily effective, and was discovered, developed and launched in time to make a difference in this pandemic. The Food and Drug Administration authorized the regimen for emergency use based on a large clinical study that clearly showed it helped keep high-risk, unvaccinated people alive and out of the hospital.

But the operative word here is “unvaccinated.” In practice today, Paxlovid is probably being taken more often by the many vaccinated people who are contracting COVID-19, as omicron and its even more infectious siblings circulate. Among the many still-unanswered questions about Paxlovid is whether vaccinated people — and especially those who have no other health risk factors — benefit from the drug.


So far, studies of the drug that have included unvaccinated, high-risk patients have provided no strong evidence that it helps people feel better faster. Early data from a Pfizer-led trial of Paxlovid in unvaccinated people who are not at high risk for serious infection and in vaccinated adults with at least one risk factor for severe illness did not reach statistical significance on the primary goal of alleviating symptoms for four consecutive days. Pfizer said last month that taking the drug prophylactically does not prevent more infections in a household where one person has tested positive for COVID-19.

Also unknown is whether Paxlovid can help prevent long COVID.

Most recently, a question has arisen about whether the drug might actually prolong infection in some patients. Certain people who take Paxlovid have found that they experience a rebound of their symptoms, and again test positive, when the pills are used up. One possible explanation is that the drug can in some cases suppress the infection enough to make the patient feel better, but not do away with it altogether.

Some infectious disease doctors wonder whether a longer course of therapy — 10 days instead of five, perhaps — might fully shut down the virus. Albert Bourla, Pfizer’s chief executive officer, has suggested that any patient experiencing a rebound should simply take a second course of the drug. That prompted the FDA to put out a statement on May 4 noting that “there is no evidence of benefit at this time” for extending or repeating treatment with Paxlovid.

The National Institutes of Health and the Centers for Disease Control and Prevention are reportedly mapping out studies to investigate the rebound infections. “It is a priority,” Clifford Lane, deputy director for clinical research at the National Institute of Allergy and Infectious Diseases, told Bloomberg News.

But given the slow pace of NIH COVID-19 studies, it seems unlikely that answers will arrive in time to guide the best use of Paxlovid in whatever variant wave arrives next.

As test-to-treat ramps up, it becomes more urgent to understand the rebound risks. Some people at low risk of COVID-19 complications might be using the drug merely to feel better faster rather than to stay out of the hospital, says Walid Gellad, associate professor of medicine and health policy at the University of Pittsburgh. If the drug isn’t fully clearing their infection and their symptoms return, Gellad wonders if the drug could actually be prolonging illness.


Doctors have other questions, too: Do new variants diminish Paxlovid’s efficacy? (Pfizer’s trials were run before the emergence of omicron.) Are these rebound infections cause to worry that the coronavirus might be developing resistance to the pills? And would the drug be more effective if it were combined with other antivirals, whether Merck’s molnupiravir or one of the newer drugs in development.

When test-to-treat was launched, an opportunity was missed to follow the progress of patients who were prescribed the pill pack. That would have provided at least some observational data on how the drug works in vaccinated people and perhaps given a signal of whether it can prevent long COVID-19.

In its first-quarter earnings call, Bourla noted that Paxlovid has been added to the U.K.’s nationwide Panoramic study to see how well it works in that highly vaccinated population.

It looks as if, once again, the U.S. will need to look to other countries for the data it needs to understand COVID-19.