Decades ago, a case report (relating the experience with a single patient) was published which described how a person's flu symptoms improved after a bowl of chicken soup, but then reappeared. The article was meant as a kind of parody of the old maxim that chicken soup is the best cure for a cold. Pediatricians occasionally see a similar phenomenon when children are treated with an antibiotic for an ear infection; they may then have to prescribe another course of the drug. We're seeing a similar phenomenon with Paxlovid, an oral two-drug combination regimen that treats COVID-19. One drug is nirmatrelvir, an antiviral, while the second, ritonavir, inhibits the metabolism of the nirmatrelvir, maintaining therapeutic levels.
Although not perfect, Paxlovid is a breakthrough, with efficacy over 90% in treating mild-to-moderate COVID-19 and preventing hospitalization and death. One drawback, however, is that the ritonavir also affects the metabolism of many other drugs, including anticoagulants, anticonvulsants, anti-arrhythmics, anti-hypertensives, and lipid-lowering drugs, which can turn a normal dose into a toxic or ineffective one. Still, it's a hugely important advance.
However, physicians are seeing an unusual phenomenon in a small number of patients treated with Paxlovid – an improvement of symptoms while taking the drug, only to have them return after completion of the five-day course. It has been dubbed "Paxlovid Rebound." There are not much data on how frequently this occurs, but Pfizer's application to the FDA for Emergency Use does state that "several subjects appeared to have a rebound in SARS-CoV-2 RNA levels around Day 10 or Day 14." And according to Dr. Mikael Dolsten, Pfizer's chief scientific officer, "We have access to more than 300,000 Paxlovid treated [patients] in [a large database]. We have reports of this happening in about 0.005% or less [of patients treated]."
Whatever the frequency, there are several possible mechanisms to explain such a phenomenon.
Has the virus become resistant to the action of Paxlovid?
The mechanism by which Paxlovid inhibits the virus (at least so far) is independent of the variants with multiple mutations in the viral spike proteins. Neither Pfizer nor the FDA is aware of any of the thousands of sequenced variants showing resistance to Paxlovid. Neither are the authors of a 2022 review, so this is probably not the correct explanation of the reported cases of COVID rebound.
Are patients being reinfected after the course of Paxlovid?
Although some people have been infected with SARS-CoV-2 more than once, especially after the appearance of Omicron, the clinical picture of people who took Paxlovid, felt better, and then experienced a reappearance of their symptoms isn't consistent with reinfection. That is because reinfection is defined as "when a person becomes infected with COVID, enough time passes, and later becomes infected again ... 90 days or more after their first positive test." The time course of the Paxlovid rebounds is too short, so it seems unlikely that people who suffered rebound COVID-19 symptoms a few days after completing Paxlovid therapy became ill again because of reinfection. Instead, they probably did not completely clear the infection with a five-day course.
Is the dose too low or the length of treatment too short to prevent rebound?
How long does it take to clear SARS-CoV-2 from your body? Ironically, the best answer to this question comes from the TOGETHER randomized controlled trial that examined the effect of ivermectin vs. placebo on multiple parameters of COVID-19, one of which was the number of days it took to clear the virus. As shown in Figure 1 here, only 25% of the participants cleared the virus after 7 days in both the ivermectin and placebo groups.
In other words, if only 25% of the trial participants were able to clear the virus in seven days with no treatment, the time to achieve complete viral clearance must be considerably longer, perhaps two to three weeks, maybe more. How long does it take for patients treated with Paxlovid to clear the virus? Although this precise information is not available from Pfizer's clinical trials, the company does provide this information: "An approximate 10-fold decrease in viral load at Day 5, relative to placebo, was observed in both EPIC-HR and EPIC-SR [trials], indicating robust activity against SARS-CoV-2 and representing the strongest viral load reduction reported to date for a COVID-19 oral antiviral agent."
This offers a solid clue about what's happening. While a five-day course of Paxlovid is sufficient to keep about 90% of COVID victims out of the hospital and reduce their viral load 10-fold, there may be a subpopulation of people who either didn't respond as well as others to Paxlovid or, perhaps, had a higher viral load to begin with.
Of the possible theories, this one makes the most sense.
In conclusion, we believe that the net benefit of Paxlovid is obvious. For those who are undecided, here is the text of a May 11 tweet by Dr. Bob Wachter, chairman of the Department of Medicine at the University of California, San Francisco, School of Medicine: "Paxlovid rebound is real & poorly understood. We need urgent study & patient counseling. But biggest risk is that the Misinformation Engine revs up, Pax gets a bad name, MDs hesitate to prescribe it & patients hesitate to take it. Pax's benefits are still much greater than its risks." Period.
Dr. Miller, a physician and molecular biologist, is a senior fellow at the Pacific Research Institute. He was the founding director of the Office of Biotechnology at the FDA. Dr. Bloom is the director of Chemical and Pharmaceutical Science at the American Council on Science and Health.