Six months after regulators issued an emergency use authorization for Paxlovid, physicians say they still have significant questions about prescribing guidelines for the leading treatment for high-risk Covid patients.
STAT spoke with providers who said they and their colleagues aren’t on the same page about when to prescribe Paxlovid or the criteria that separates those who need it from those who do not. They also said it is unclear whether they can give a second course when patients test positive again after taking Paxlovid, a phenomenon known as a rebound. And nearly all the experts who spoke with STAT said that they are clamoring for more data on rebounds, which is complicating and sometimes changing their calculus about when to give the drug.
“There is a real dearth of evidence right now out there, and obviously there’s a lot of confusion,” said Jonathan Li, a physician at Brigham and Women’s Hospital and virology researcher at Harvard Medical School who is also a member of the Covid-19 Treatment Guidelines Panel. “And even amongst people who are immersed in the literature, and who are infectious disease experts, you’ll see actually a fairly wide range in opinions.”
The pool of people who can prescribe Paxlovid is now even broader — on Wednesday, the Food and Drug Administration began allowing pharmacists to prescribe the drug, which was shown to reduce the risk of hospitalization and death by 89% in an early study of unvaccinated patients. The FDA says Paxlovid — a combination of two drugs taken twice a day for five days — should be prescribed within five days of symptom onset to people 12 and older who have tested positive and are “at high risk for progression to severe Covid-19.” Technically, that includes everyone from people who have asthma or obesity to people who are over age 65.
“Pretty much everybody meets the EUA criteria. They made it very, very, very broad,” said David Smith, a professor, physician, and virology researcher at the University of California, San Diego. In a single week in May, over 160,000 Paxlovid prescriptions were filled.
Physicians generally agree that certain high-risk patients — including people who are unvaccinated or those over 65 with multiple comorbidities — should always be prescribed the drug. But the broader eligibility makes it difficult for some physicians to decide who should or should not receive Paxlovid. A child is not likely to need it, but what about a healthy 50-year-old man? A 65-year-old woman? The experts STAT spoke with didn’t agree.
Debra Poutsiaka, interim chief of the Division of Geographic Medicine and Infectious Diseases at Tufts Medical Center, recommends Paxlovid to all patients who qualify. “There’s a reason why the EUA recommends people of certain groups — meaning comorbidities and age — to get Paxlovid,” said Poutsiaka. “Studies, over and over again, have shown that people falling within those criteria are at increased risk for severe Covid. So I wouldn’t say or make my own judgment about whether or not I think someone should get Paxlovid.”
Some physicians are more conservative when recommending Paxlovid, though they still encourage older patients with one or more comorbidities to accept it. For them, it’s the borderline cases — like patients who are older, but otherwise in good health — that are tricky. The disease can manifest in many different ways in different patients, and Paxlovid, like any drug, does come with potential side effects.
“It’s a nuanced decision when it comes to lower-risk people,” said Rajesh Gandhi, an infectious diseases physician at Massachusetts General Hospital and Harvard Medical School. “Some of this does come down to clinical judgment on a case-by-case determination.”
Differences have also emerged in the timing of prescribing. Myron Cohen, an infectious diseases specialist and prominent HIV researcher at the University of North Carolina at Chapel Hill, advocates starting on Paxlovid immediately. “The conversation’s a pretty short conversation,” said Cohen. “The data that’s available suggests that if you want to have the maximum benefit of this drug and you want to try to reduce long Covid, that the sooner you take it after you test positive, the smarter you probably are.”
Smith, the UCSD professor, said he’s seen physicians prescribe the drug after a patient is exposed but before they test positive, “even though the study is fairly clear it didn’t have a benefit as prophylaxis.”
Conversely, Robert Wachter, chair of the Department of Medicine at the University of California, San Francisco, speculated that giving Paxlovid early may not give the immune system enough time to ramp up and prepare itself for when the drug tapers off, which he worries may increase the risk of a rebound. (There is no research yet to demonstrate this is the case.) “For someone that I think of as being a borderline candidate, I would say the majority outcome is — if it’s on day one or two — is let’s wait three. … We have till day five to take this,” said Wachter. “Let’s wait until day four and see how you’re doing and if you’re feeling well, don’t take it. If you’re still feeling really bad, then it’s reasonable to take it.” Of Wachter’s patients who followed his advice, some did end up taking the drug. But for “the majority of people,” symptoms were minimal by day three and Paxlovid was not necessary.
Wachter said he’d follow the same advice he gives patients. “I’m 64, I’ve had two boosters — two vaccines and two boosters,” said Wachter. “I think if I got Covid today, I would still take Paxlovid. And would I wait a day or two? I wouldn’t rush to get it on day one. I might wait for day two or three.”
But other physicians disagree about waiting. The first and largest study of Paxlovid treated patients within three days of their experiencing symptoms, so the five days in the emergency authorization already offers leeway, they say, that isn’t as well-supported by the scientific evidence. “Our knowledge is greater about early use than late use. So that kind of ‘let’s see how you do’ doesn’t really make sense to me except in people in whom you don’t think they need Paxlovid,” explained Cohen, the UNC-Chapel Hill physician. Cohen said he and colleagues in his practice prescribe as soon as possible to patients who they believe need Paxlovid.
Cohen and other physicians also have to contend with Paxlovid rebound, which occurred in one to two percent of patients in the initial clinical trial that resulted in the FDA approval of Paxlovid. Some experts said, anecdotally, they’ve found rebounds — which generally occur four or five days after treatment with Paxlovid — to be much more common.
Scientists don’t yet have a clear understanding of why rebounds happen, or how prevalent they are. That has physicians uncertain about how much to take them into account when prescribing.
“I think the sense of likely risk-benefit has changed over time,” said Walid Gellad, an internal medicine physician and health policy researcher at the University of Pittsburgh. “So I have a little more reluctance now than I did at the very beginning on prescribing.”
Gellad and other experts said rebounds warrant more urgent research. “There hasn’t been, I think, enough attention to asking the question of ‘Why do we not know?’” he said. “Why does the company not know at this point how common rebound is? Why is the FDA not insisting that we know more about rebound?”
In perhaps the highest-profile case of a Covid-19 rebound, Anthony Fauci, the Biden administration’s chief medical adviser, recently experienced rebound after a first course of Paxlovid and received a second regimen of the drug as a result. It’s an unusual move, and one that had experts conflicted. None of the physicians STAT spoke with have prescribed Paxlovid for a patient experiencing rebound, primarily because rebound cases often resolve on their own and rarely progress to a hospitalization or death. But it was unclear to them whether the emergency use authorization allowed a second course, with some physicians stating this was up to each individual state’s medical board and others arguing both sides.
“Some pharmacists have been pretty assiduous,” said Cohen. “When I tried on occasion to give a second course of Paxlovid over a short window of time, they basically argued with me and refused it, which is an unusual thing with a pharmacist.” Li, the Brigham and Women’s Hospital physician, pointed out that the wording of the authorization was unclear as to whether a rebound qualified as a new “day zero” of symptom onset.
In a statement to STAT, the American Academy of Family Physicians cautioned “against prescribing medications for any off-label use that is not supported by significant scientific evidence” and pointed to a Centers for Disease Control and Prevention health advisory that said “there is currently no evidence that additional treatment is needed” for Covid-19 rebound. The FDA and CDC did not respond to questions about Paxlovid prescriptions.
Ultimately, the confusion around Paxlovid, a pill largely tested during the Delta wave, is part of a broader discussion about how best to treat patients as the virus evolves. When the Omicron variant arrived, existing monoclonal antibodies that had once been a primary line of defense were found to be less effective at preventing infection. In response, the FDA doubled the dosing of AstraZeneca’s Evusheld but curtailed the use of Eli Lilly’s and Regeneron’s antibodies. Federal authorities no longer recommend the other preventative treatments.
At least one physician is looking beyond the current discussion, to a time where Paxlovid or another drug would have full approval and be widely used. Paul Sax, clinical director of the infectious diseases division at Brigham and Women’s Hospital, said he could see a future in which “we end up managing Covid-19 much as we manage influenza, which is that we do prescribe antiviral therapy to many people who have flu if they seek care early enough.”
For now, physicians and the existing scientific studies support Paxlovid treatment for patients with a high risk of developing severe disease. And rebound or not, it is hospitalizations and deaths that matter. “I tell my patients this, that in the end, that is the most critical outcome and that it shouldn’t dissuade us from using Paxlovid in those patients who are at high risk of severe disease,” said Li. “Because we know it works.”