select search filters
roundups & rapid reactions
before the headlines
Fiona fox's blog

expert reaction to study looking at molnupiravir and new SARS-CoV-2 mutational signatures

A study published in Nature looks at molnupiravir-associated SARS-CoV-2 mutational signatures. 


Dr Chris Illingworth, Senior Lecturer (Virology), University of Glasgow, said:

“This is an important and careful study, showing that the use of molnupiravir to treat COVID infection is associated with the transmission of more mutated, and differently mutated viruses.  However, there is a need for caution before simply abandoning the use of the drug.  Most mutations would be expected to make the virus less, rather than more dangerous.  As such we can’t yet say that the use of molnupiravir has been harmful for human health.”


Prof Chris Butler FMedSci, Professor of Primary Care, Clinical Director of the Primary Care Clinical Trials Unit, and NIHR Senior Investigator, University of Oxford, said:

“This large-scale, international observational study has identified potentially concerning associations between molnupiravir use and worrying mutations: As with the parallel pandemic of antibiotic resistance, these findings underline the importance of doing rigorous research into possible changes in the make up of microbes as a consequence of widespread antimicrobial use.  We must not make the same mistakes we made with antibiotics, when we used them at scale before doing trials to find out who really benefits and what the consequences might be when it comes to antibiotic resistance.  No antimicrobial should be used at scale until we have rigorous clinical trial evidence about who will benefit, and who can safely be managed without antimicrobial treatment.  Certainly, molnupiravir should not be used by all who get infected with the SARS-CoV-2 virus.  However, in some, pressured circumstances the drug could be very useful in promoting systems resilience, as it clearly helps people get better quicker and helps reduce viral load.”


Prof Jonathan Ball, Professor of Molecular Virology, University of Nottingham, said:

“This is really interesting data and shows a strong link between molnupiravir treatment and the subsequent emergence, and occasionally the limited spread, of SARS2 coronaviruses with highly mutated genomes – genome changes that carry genetic signatures that indicate their emergence was likely caused by the drug itself.  The emergence of viruses with such highly mutated genomes is not surprising – the drug works by increasing the mutation rate of the virus to a level whereby offspring viruses contain so many mutations that the resulting virus is not viable, a process called ‘error-catastrophe’.  What is surprising though is that some of these viruses were able to transmit to others, showing that their mutation spectrum wasn’t lethal after all.  But what isn’t clear is if any of the transmitted viruses contained mutations which would change how they would behave – for example if they were more or less transmissible, more pathogenic or less susceptible to our immunity, and this is a key question to address.”


Prof Stephen Griffin, Professor of Cancer Virology, University of Leeds, said:

“This paper is an incredibly important, well-conducted piece of research.  The findings are of critical value to our understanding of how the use of this specific antiviral drug could have been better implemented, but also reminds us of more general aspects of good practice and antimicrobial stewardship.

“It is worth noting that the use of this drug is not immediately dangerous to individuals taking it, but these findings have important implications for the future direction of the pandemic.

“Molnupiravir has long been known to be effective against SARS-CoV2 and related coronaviruses in the lab, targeting the viral enzyme that replicates its RNA genome.  However, it does this in a different way to e.g. remdesivir.

“Instead of tricking the enzyme, known as a polymerase, into adding a link in the growing RNA chain that can’t have another link attached, it instead becomes part of that chain, replacing one of the normal building blocks.  Then, when the polymerase comes back again, copying another new chain of RNA, it misreads the molnupiravir building block, and makes a mistake.  As more drug building blocks become incorporated, the more errors occur, and eventually the RNA being built carries multiple errors and is next to useless as it no longer carries the right message to make functional proteins.

“This process is referred to as “error catastrophe”, but it may also be that errors occur in vital areas of the RNA that immediately prevent it from being viable, so-called “lethal mutagenesis”.  Other antivirals like ribavirin, used to treat hepatitis C virus, and favipiravir, used in Japan vs influenza, are thought to act this way too.

“The way this process happens means that the activity of molnupiravir leaves a defined “signature mutation” that would otherwise only occur very rarely in the virus.  Of course, if the drug does its job, this shouldn’t matter as the virus within a successfully treated patient won’t be going anywhere.

“However, the researchers behind this study found that molnupiravir signatures could indeed be detected in multiple virus sequences in different countries, and this coincided with when the medicine began to be used as a mainstay of antiviral therapy.

“What was troubling though, was that frequently the viruses with these signatures tended to have accumulated many such mutations, as well as other mutations often seen in variants of concern such as delta and omicron.

“This pattern usually occurs in patients with long term SARS-COV-2 infection, often where the immune system is unable to fight the virus off, and multiple rounds of antivirals are needed to treat them.  The fact that these mutated viral lineages have been detected means that, rather than killing the virus off, molnupiravir in fact contributed to the evolution of a virus that was then able to spread, i.e. it didn’t induce either fatal errors OR a catastrophic number, that would make that virus non-viable.

“It’s important to stress that this won’t happen in the majority of people treated with this drug.  However, as we have seen recently with BA2.86, SARS-COV-2 evolution sometimes takes “leaps”, and anything that might help that occur is a bad idea.

“Molnupiravir is currently being used less and less as its efficacy when used alone in vaccinated, non vulnerable patients, is poor in terms of preventing people going into hospital.  It does help in at-risk non vaccinated patients, but only at ~ 30% efficacy, contrasting remdesivir and paxlovid at >90% when given early enough.

“So, does this mean we should stop using molnupiravir?  Well, taking this new evidence together with evidence from the PANORAMIC trial, there is the suggestion that we should think about whether molnupiravir should be used on its own – but it shouldn’t be discarded and could still be valuable if we were to use it in drug combinations.

“Logically, the way molnupiravir works explains its reduced potency compared to remdesivir or nirmatrelvir (the active component of paxlovid), as the virus can clearly tolerate exposure to the drug and ensuing mutagenesis to a certain extent, and so possibly survive the short 5 day course given to patients in some cases.

“But, if combined with another antiviral such as nirmatrelvir, which immediately poisons the viral protease, it may be that the sum of the two is greater than its parts, i.e. they could act synergistically.  Drug combinations have proven effective when treating immunosuppressed patients with persistent SARS-COV-2 infection, but so far no trials have been set up.  Drug combinations would also reduce the chance of molnupiravir mutation signatures occurring in transmissible viruses.

“The other benefit would be to reduce the likelihood of drug resistance emerging.  The beginning of this is already occurring, due mainly to the use of a single drug against such a variable virus.  We know this is a bad idea from HIV, hepatitis C virus, and indeed, influenza.

“Thus, it will be important to reassess how antivirals are deployed in the future.  This is especially important as current COVID policy leaves vulnerable people exposed, and so dependent upon access to prevent severe COVID and hospitalisations.  Pharmaceutical companies need incentives to cooperate and set up combined trials, which will ultimately extend the usefulness of these essential drugs going forwards.

“What remains certain though, is that we must remain vigilant to the ongoing shifts and leaps in SARS-COV-2 evolution, which is becoming all the more difficult to track due to reductions in testing and surveillance.  The authors of this excellent paper benefitted from access to huge amounts of sequencing data when searching for these mutational signatures.  It may be that going forward this will become more and more difficult to achieve.”


‘A molnupiravir-associated mutational signature in global SARS-CoV-2 genomes’ by Theo Sanderson et al. was published in Nature at 16:00 UK time on Monday 25 September 2023. 

DOI: 10.1038/s41586-023-06649-6 (2023)


Declared interests

Prof Jonathan Ball: “No CoI.”

Prof Stephen Griffin: “Co-Chair of Independent SAGE.”

For all other experts, no reply to our request for DOIs was received.

in this section

filter RoundUps by year

search by tag