Yesterday I wrote about the potential dangers the antiviral drug molnupiravir could unleash by supercharging new SARS-CoV-2 variants. Today, my focus is on the people who may receive the drug as a treatment and the possibility that molnupiravir could lead to cancerous tumors in those patients and birth defects in the unborn.

Molnupiravir is a relatively new drug, initially developed as an antiviral treatment for influenza. Molnupiravir’s metabolite, an active compound called NHC, has been known and studied for decades. The metabolite works by creating havoc with RNA polymerase, the enzyme critical for viral replication. As I described in yesterday’s article, the drug inserts errors into the virus’ genetic code every time it copies itself. Insert enough errors and you essentially kill off the virus, preventing it from replicating any further. 

Against other coronaviruses, like MERS-CoV and mouse hepatitis virus (MHV), the drug was found to create up to more than a hundred mutations at every section of the viral genome. Against SARS-CoV-2, molnupiravir’s manufacturers Merck and Ridgeback say that the drug’s antiviral effects are powerfully effective, limiting the virus’ ability to proliferate unchecked and cutting the risk of hospitalization and death by half among those infected. The trouble with the drug, however, is that its mutagenic powers may also create havoc among other enzymes in the body, including the nucleic acids in our own healthy DNA. 

As far back as 1980, researchers have been trying to understand just how damaging NHC, molnupiravir’s metabolite, can be to our own healthy cells. Earlier this year, a study published in the Journal of Infectious Diseases found that the metabolite could indeed be incorporated into and mutate within our host DNA. As others have pointed out, just because something is mutagenic doesn’t mean it’s entirely bad — even sunlight is mutagenic. But, just like sunlight, overexposure can lead to long term ill effects, like cancer. In the case of molnupiravir, the drug may not just lead to the growth of cancerous tumours but also, potentially, to birth defects, either through sperm precursor cells or in pregnant women.

Molnupiravir has been tested for mutagenicity in animals before being moved to human trials, where it is being tested for safety. But that doesn’t mean the drug is fully in the clear. The pool of participants in the clinical trial — around 1,500 patients — is too small to pick up on rare mutagenic events and the early nature of the trial is too short-term to provide a proper view of issues that may occur months, if not years, down the road. Merck would do well to remember their experience with Vioxx, a painkiller that was deemed safe based on initial studies, but later proved deadly. The FDA originally approved Vioxx based on a safety database that included around 5000 people. Five years later, the drug was recalled after a broader and longer term study found a definitive link between the drug and rare cardiac events. There is evidence that during the time the drug was on the market it may have killed up to 56,000 people and left up to 140,000 with heart disease. 

I believe Merck and Ridgeback know there are questions around the possible mutagenicity and teratogenicity of molnupiravir that need to be answered. Both male and female participants in the trial were asked to abstain from sex or use contraception during and shortly after the trial. And reporters have asked the manufacturers about potential mutagenic effects, which Merck has answered by saying that, “the drug will be safe if used as intended and at the concentrations where we have looked and in the concentrations which we are achieving in patients.” 

This isn’t the first time a mutagenic drug has been tested for antiviral activity. In that respect, our prior experience with another antiviral drug, favipiravir, may be of interest. It too is an antiviral that targets RNA polymerase, initially developed as a treatment for influenza and, like molnupiravir, now being tested against SARS-CoV-2. The two drugs work in a similar fashion, interchanging two of the four letters of the viral RNA code to create copying errors — molnupiravir switches uracil (U) and cytosine (C), while favipiravir switches guanosine (G) and adenosine (A). Like molnupiravir, favipiravir works by creating enough copying errors during replication to essentially kill off the virus.

The study on mutagenicity of the molnupiravir metabolite in the Journal of Infectious Diseases earlier this year also tested favipirivir. The study found that the molnupiravir metabolite, NHC, was a far more potent mutagen than favipirivir (FAV) (see Figure 1), which is a drug that has widely known issues related to teratogenicity and links to birth defects.

Graph showing genotixicity of molnupiravir metabolite, NHC.
Figure 1. The left side of the panel shows the genotixicity of molnupiravir metabolite, NHC, in hamster cells at increasing concentrations and compared to UV light and the control vehicle. The right side of the panel shows the genotoxicity ofSHUNTAI ET AL., THE JOURNAL OF INFECTIOUS DISEASES, VOLUME 224, ISSUE 3, 1 AUGUST 2021, PAGES 415–419

Because of this, favipiravir has not been approved in the US or the UK, and is only approved in Japan under the strictest of regulations and for the most severe form of influenza, for which no other drugs exist. The challenge with approval of any drug, even under the strictest of regulations, is that once approved it can be used for many purposes, off-label. Favipiravir is already reportedly being distributed and used in Hungary as a treatment for Covid-19, despite no clear evidence of its value and its known risks. This is a danger not just to those receiving the drug, but also — as I wrote about in my previous article on molnupiravir — a danger to all of us, given the potential of drugs like these to supercharge the creation of viral variants.

In November, the FDA will debate the use of molnupiravir only as a treatment for high-risk individuals with mild to moderate disease. But an initial emergency use approval for the drug may lead to unknown harm to all those who receive it. Yesterday, I wrote that if the FDA approves the drug it should be only on a very narrow basis and include a black box warning to emphasize the potential danger of using the drug at suboptimal doses or for large numbers of people for preventive purposes. To that, I add the need for specific warnings for men and women who are actively trying to become pregnant and for women who are already pregnant — under no circumstance should these individuals receive this treatment. I hope that Merck, Ridgeback, the FDA and the CDC explore the dangers of molnupiravir thoroughly before granting emergency use approval of the drug, especially as other potentially safer antivirals are already on the way. This drug could harm the very people it’s meant to help, those receiving the drug and all of us around them, should new more powerful variants be unleashed.

* Shuntai et al., The Journal of Infectious Diseases, Volume 224, Issue 3, 1 August 2021, Pages 415–419, full caption: HPRT assay to detect genotoxicity of rNHC, RBV, FAV, AZT, 3TC, and TDF in CHO-K1 cells. A, 6-thioguanine-resistant colony counts in 2 separate HPRT mutagenesis experiments. In the HPRT experiment 2, an additional round of initial cleansing for spontaneous HPRT mutations was conducted to limit background mutations. Each compound/dose group had 3 replicates. Average numbers of colonies are shown on the top of each bar. Significance compared to vehicle control (* P = .01–.05, ** P = .001–.01, **** P <.0001) was determined using the unpaired t test calculated using the GraphPad Prism version 8.3.0 built-in function.