Almost lost in last week’s blockbuster release of Omicron: Fear and Loathing — The Rise of the Next Variant was an FDA panel’s recommendation that a drug named molnupiravir be given Emergency Use Authorization (EUA).
The drug would be the first of its kind, a treatment for COVID that can be taken at home, as a pill. Its most well-recognized “relative” might be the anti-influenza pill Tamiflu (oseltamivir), which works by blocking the influenza virus’ ability to exit an infected cell. If taken within 48 hours of the onset of symptoms, Tamiflu has been shown to reduce the intensity and complications of the flu, and perhaps even death.
Molnupiravir (Lagevrio) works a little differently. The virus mistakenly inserts a metabolite of the drug into its RNA — as if an extra number or symbol were inserted into one of your computer passwords. The substitution causes the virus to produce mutated versions of itself, calamitously referred to as “viral error catastrophe.” With no meaningful viral replication, the flame goes out.
Does it work?
Merck, the manufacturer of molnupiravir, presented the FDA with results of their MOVe-OUT study, which randomized participants to receive either molnupiravir or a placebo. To enroll, participants had to be unvaccinated and have a laboratory-confirmed COVID-19 infection that was either mild (flu symptoms but no shortness of breath, and a normal chest X-ray) or moderate (shortness of breath, abnormal chest X-ray, but maintaining oxygen levels on room air).
They also needed to be early in the disease course (less than five days) and had to have at least one risk factor for progressing to a serious case of COVID. In this study, obesity (77%), age over 60 years (14%), and diabetes (14%) were the most common risk factors.
In October, Merck released data on the first half of the 1,400+ enrollees that had completed the study. Just 7% of those who took molnupiravir ended up being either hospitalized, or, well, dead, compared with 14% of those who received the placebo. For the mathematically challenged, that’s a 7% absolute risk reduction and a snappy 50% relative risk reduction. (A word of caution: relative risk can overstate benefit when the absolute risk is small. For instance, you may be relieved to know that the crocheted beer can hats I sell on Etsy can reduce one’s risk of being struck by an asteroid by 50%. It goes from 2 in a billion to 1 in a billion.)
But when Merck released the final results of the study last week, the numbers were less snappy. The molnupiravir group held at the 7% mark for hospitalization or death, but the placebo group did better than the first half results suggested, dropping from 14% to 10%. In this study, hospitalization and death were lumped together as a single outcome, but of course they are drastically different outcomes. For the record, there were nine deaths in the placebo group and one in those taking molnupiravir.
Molnupiravir didn’t benefit the vaccinated. But will the unvaccinated want to take it?
It’s important to remember that none of the patients in this drug trial had been vaccinated, but that doesn’t mean they hadn’t been in contact with the virus itself. When researchers looked at the small percentage of trial participants who happened to have antibody testing results available, molnupiravir wasn’t of any benefit for those who already had antibodies to COVID. (Trials for monoclonal antibody therapy on the other hand seemed to show benefit for those with or without prior antibodies.)
This raises an obvious question that has no clear answer. If someone has refused a vaccine, how likely would they be to take molnupiravir? To try to answer that, let’s look at a couple of safety issues that researchers and the FDA advisory committee reviewed.
Could molnupiravir create a new variant?
The term “viral error catastrophe” has a nice ring to it (“Die COVID, die!”), but could molnupiravir’s mechanism of action — causing RNA mutations — actually backfire?
RNA viruses are notorious bunglers when it comes to replicating, but that’s also part of their genius: the bungling leads to new mutations. Most of those mutations will be duds, but the rare “winners” could go on to create new variants. Omicron has twice the number of mutations in its spike protein that delta has, but some of those mutations might be incidental and of no biologic consequence.
Researchers performed a viral genetic analysis on a small number of trial participants to see if any new mutations developed during treatment. Although those taking placebo showed spontaneous (“bungling”) mutations in the particular COVID strain that had infected them, taking molnupiravir was associated with a “modest but significantly higher” mutation rate.
What does that mean in real life, outside the lab? To summarize the FDA advisory committee’s briefing document: It’s hard to know. They noted that viral shedding will likely be rapidly declining by the time a molnupiravir-associated variant could emerge in the host, and if the host can’t spread it, the variant’s success story ends right there.
“Unlikely” does not mean “impossible,” though. Even if the likelihood of molnupiravir triggering a mutation is extremely low — let’s just say one in a million — if the drug ends up being prescribed to millions, well, then, new variants are probably going to happen. According to CNN, the U.S. federal government has already contracted to buy 3.1 million courses of molnupiravir.
Could taking the drug lead to mutations in our own cells?
We may think of DNA as the sacred recipe book, but to make a meal you need a chef — and RNA is what will read the recipe and get the cooking started. Molnupiravir is designed to interfere with viral RNA production only, but the FDA advisory committee noted the theoretical possibility that molnupiravir could be taken up and incorporated into our DNA as a mutation.
The committee found evidence that molnupiravir caused some DNA mutations in bacteria, but not in mammals. (Fun fact: humans have enzymes to repair DNA mistakes/mutations, but bacteria do not.) Overall, the committee described the likelihood of permanent alterations to the recipe book as being low.
Molnupiravir was well tolerated, but because animal studies suggest that molnupiravir interferes with bone and cartilage formation, and fetal development as well, the committee recommended that the drug not be given to patients less than age 18, or to pregnant women without a serious consideration of risks and benefits.
The approval was narrow
If you’ve read this far and have misgivings about molnupiravir, apparently the FDA advisory committee felt the same way. They voted 13 for and 10 against recommending it. Pfizer has an oral anti-viral drug named paxlovid that works via a different mechanism than molnupiravir. They’ve already submitted it to the FDA for Emergency Use Authorization, and we are all hoping that paxlovid’s performance and safety profile will warrant a more enthusiastic approval vote. Stay tuned.