Prescriptions for Pfizer’s blockbuster drug Paxlovid have skyrocketed in recent weeks. That’s good news for many COVID-19 patients, as the pill has been proven to reduce severe disease from SARS-CoV-2 infections. But a bevy of new lab studies shows the coronavirus can mutate in ways that make it less susceptible to the drug, by far the most widely used of the two oral antiviral drugs authorized to treat COVID-19 in the United States. Researchers have found some of those mutations in variants already circulating in infected people, raising fresh concerns that physicians could soon lose one of their best therapies for fighting COVID-19.
Taken together, the studies show that “when you put pressure on the virus it escapes,” says David Ho, a virologist at Columbia University who was among the first to document drug resistance mutations in HIV some 3 decades ago. Ho was not involved with the new studies but is conducting similar work on SARS-CoV-2. Although the newly identified mutations have yet to become widespread, Ho and many other scientists think it’s only a matter of time. “Given the amount of infections out there, it’s going to come,” Ho says.
The resistance studies come on the heels of other recent concerns about Paxlovid, which in the United States remains restricted to use in people with risk factors making them more likely to develop severe COVID-19. Confirming anecdotal reports widely reported by media, several studies have found a small percentage of infected people who receive the normal 5-day course initially feel better, only to have their symptoms rebound. And questions have grown about whether Paxlovid helps those who aren’t at high risk of serious disease—Pfizer earlier this month halted a large trial of the drug in standard risk COVID-19 patients because it was failing to show statistically significant protection against death or hospitalization.
The U.S. Food and Drug Administration (FDA) granted emergency use authorization for Paxlovid in December 2021. The drug consists of nirmatrelvir, the active antiviral, and ritonavir, a compound that slows the breakdown of nirmatrelvir in the body. Because of bottlenecks in manufacturing nirmatrelvir, Paxlovid’s rollout was slow—doctors in the United States issued only 40,000 or fewer prescriptions per week through mid-April. Since then, prescriptions have surged to more than 160,000 per week, according to the latest numbers from the Centers for Disease Control and Prevention.
That rise creates selective pressure on the virus, favoring mutations that help it survive in the presence of the drug. And because each infected person makes trillions of copies of SARS-CoV-2, the virus has plenty of opportunities to test out different mutations as it replicates.
So far, those mutations don’t seem to have interfered with Paxlovid’s effectiveness. Nirmatrelvir prevents SARS-CoV-2’s main protease (MPRO) from cutting a long precursor molecule made by the virus into shorter active proteins, an essential step in SARS-CoV-2’s reproduction. In February, Pfizer researchers reported in JBC Accelerated Communications that nirmatrelvir remained effective in halting the activity of MPRO in multiple SARS-CoV-2 variants, including Alpha, Beta, Delta, Gamma, Lambda, and Omicron, as well as the original strain.
However, the recent studies suggest the virus is poised to develop resistance—a fate that befalls many antiviral drugs. Two preprints posted on bioRxiv on 7 June, for example, show that SARS-coV-2 grown in the lab quickly gains the ability to avoid nirmatrelvir’s attack. Two research groups independently cultured the coronavirus with low levels of nirmatrelvir, killing some but not all of the virus. Such tests are meant to simulate what might happen in an infected person who doesn’t take the whole regimen of the drug or an immunocompromised patient who has trouble clearing the virus.
One of those studies, led by Dirk Jochmans, a virologist at KU Leuven in Belgium, found that after 12 rounds of nirmatrelvir treatment, SARS-CoV-2 accumulated three mutations—at positions 50, 166, and 167 in the string of amino acids that make up MPRO–that reduced the virus’ susceptibility to nirmatrelvir 20-fold, as determined by the dose of drug required to kill half the virus in a sample. The other study, led by Judith Margarete Gottwein, an immunologist at the University of Copenhagen, also spotted potential resistance-conferring mutations at positions 50 and 166 in MPRO . When those mutations occurred together, the virus was 80 times less susceptible to nirmatrelvir. “This tells us what mutations we should be looking for [in patients],” Gottwein says.
Indeed, some of these mutations are already in coronavirus-infected people, according to work by Adam Godzik, a bioinformatics expert at the University of California, Riverside. Godzik and his colleagues scoured the GISAID database, a catalog of more than 10 million SARS-CoV-2 genomes sequenced from viruses isolated from infected individuals, searching for amino acid changes at positions in MPRO near where nirmatrelvir binds. In a bioRxiv preprint posted on 30 May, they reported that mutations to amino acids 166 and 167—two of the resistance mutations flagged by the Belgian group—were already in viruses circulating in people. Because these mutations occurred before widespread use of Paxlovid, they likely occurred randomly, Godzik says. However, he adds, they reveal the enzyme has some flexibility at these positions that could help the virus work around the drug.
And the list of potential resistance mutations keeps growing. In a paper posted yesterday on bioRxiv, Jun Wang, a medicinal chemist at Rutgers University and colleagues report 66 common mutations to MPRO near the nirmatrelvir binding site. Like Godzik’s team, they scanned the GISAID database to find altered versions of the protease, but then went a step further. Adding the gene for each of these variants of MPRO to Escherichia coli bacteria, they created supplies of the enzymes for additional tests: first to determine whether each variant still carried out the essential duties of cutting viral proteins, and second to determine whether the mutations allowed MPRO to resist nirmatrelvir. Eleven of the 66 variants retained the protease’s function (the others impaired it), and five of the 11 were resistant to nirmatrelvir, requiring at least a 10-fold increase in the drug to kill half the virus in the sample. One of those variants had a previously seen resistance mutation, at position 166, but the other four had novel workarounds at positions 144, 165, 172, and 192. The bottom line from all this work, Wang says: “It’s just a matter of time before we see resistance emerge.”
So, why hasn’t it happened already? One possibility is that not enough people have taken Paxlovid yet to force the virus to mutate. Another explanation, Wang says, is that it may take multiple mutations in MPRO for the virus to get around Paxlovid while remaining both fully functional and easily transmissible. Thus far, adds Aditya Shah, an infectious disease specialist at the Mayo Clinic, studies show that patients who have rebound of symptoms, which happens in just 2% or fewer of those who take the drug, the rebound does not seem to be due to resistance mutations. “It’s reassuring,” Shah says, but not proof the virus won’t eventually find its way around the drug.
Pfizer says its Paxlovid regimen may forestall resistance. Patients only take the drug for a short period and typically get a dose “manyfold higher” than that required to prevent the virus from replicating in cells, thereby minimizing the opportunities for the virus to mutate, says Kit Longley, a company spokesperson.
Giving patients multiple antivirals could help prevent resistance by making it harder for the virus to evolve its way around different compounds at the same time, a strategy that has proved highly effective in treating other viruses, including HIV and hepatitis C, Ho says. Two other SARS-CoV-2 antivirals are authorized in the United States, but they have drawbacks. The other oral drug, molnupiravir, has proven considerably less effective than Paxlovid, and has raised safety concerns because it induces random genetic mutations in the virus—that typically stops it from replicating but could also spawn dangerous new variants, some scientists caution. And remdesivir, which interferes with the ability of the virus to copy its genome, is only authorized for hospitalized patients and must be delivered intravenously. A preprint posted on bioRxiv yesterday suggests combining molnupiravir and nirmatrelvir is more effective in combating SARS-CoV-2 infections than either antiviral given alone, at least in mice. But the strategy has yet to be widely embraced by doctors.
Meanwhile, pharmaceutical companies are racing to complete clinical trials on additional SARS-CoV-2 antivirals, some targeting MPRO at different sites. But those aren’t available yet. And numerous researchers, including representatives of the nonprofit Drugs for Neglected Diseases Initiative, have complained that Pfizer has not made Paxlovid easily available for trials of combination therapies. The company has said it was planning on doing those studies itself, although some are skeptical.
Until more antiviral drugs become available, Paxlovid will remain essentially alone, raising fears that sooner or later it will lose its punch. When pressed by a single antiviral, viruses usually find a way around the drug, Gottwein says. “If it can happen it will happen.” And at least according to the latest lab results, it can happen.