Peace, love, Paxlovid and Pfizermectin

Epidemiology &c.

As the new Omicron variant of concern of COVID-19 emerged over Thanksgiving week (of all times!), attention has been turning to Paxlovid (PF-07321332). If, like me, you torment yourself by perusing conspiracy theories, you might have heard it referred to as Pfizermectin. Behind that is the idea that the “COVID pill” Paxlovid is, basically, a re-packaging of ivermectin, a disproven treatment for COVID-19.[1]

UPDATE: on 03 December 2021, the USAN Council approved the name nirmatrelvir for PF-07321332.

The Paxlovid and Pfizermectin argument

The Pfizermectin argument rests on the fact that Paxlovid, an oral treatment for COVID-19 (colloquially known as a “Covid pill”) is a protease inhibitor, and so is ivermectin. Consequently, so the reasoning goes, ivermectin is just as good — except it is no longer an economically viable drug, vis-a-vis Paxlovid. So, it is claimed, ivermectin was repackaged, and now sold at a premium.

So suggests, for instance, paleo-libertarian Lew Rockwell:

Many bring up the use of ivermectin in India to ‘prove’ this point:

The problem is… Paxlovid is not ivermectin. It is not even similar, and the difference is crucially important to why Paxlovid is likely to be an effective medication, and ivermectin won’t.

But Paxlovid is not Pfizermectin at all

Paxlovid (PF-07321332, in blue) bound to 3CLPro (volume model). Author’s own work, based on PDB:7SI9.

As always, there’s a kernel of truth in the overall lie. It is true that both Paxlovid and ivermectin are inhibitors of the 3-chymotrypsin like protease (3CLPro). 3CLPro is the principal protease of coronaviruses. It belongs to a wider class called mixed-nucleophile proteases superfamily A. Their main function is to find peptide bonds in proteins between glutamine on one hand and serine, glycine or alanine on the other.

To understand what’s going on here, we need to look at protein synthesis by viruses. A lot of the time, viral proteins are produced as one big, long protein called a polyprotein. Then, viral proteases break these at pre-determined cleavage sites into the constituent parts of what eventually becomes a viral particle. As far as the virus is concerned, interference with that cleavage is bad, bad news, essentially inhibiting the ability of the virus to create copies of itself that are effective infectious agents.

This ‘molecular scissor’ in coronaviruses is 3CLPro, and as such, it’s clearly a target of interest. Mody et al., in a paper that has earned its spot on my list of top 10 papers of 2021, looked at 3CLPro inhibitors, and found that around 50 µM (micromoles) of ivermectin were sufficient to achieve a 75% or so reduction in the activity of 3CLPro.[2]

However, that’s actually bad news — because 50 µM is, actually, a lot. And therein lies the problem.

The Pfizermectin claim fails because it ignores the pharmacokinetics of Paxlovid

The problem with ivermectin is very simple. In order for an enzyme inhibitor to be effective, it needs to reach a concentration where it inhibits the target enzyme enough to make a dent in the underlying process. The standard metric for this is known as IC50 — the concentration of the inhibitor required to inhibit enzyme activity by 50%. This value for ivermectin is somewhere in the realm of tens of µM for 3CLPro.

One µM of ivermectin weighs around 0.0875mg, so a 1 µM solution of ivermectin would have a concentration of around 87.5 nanograms per millilitre (ng/mL). To reach even a very conservative 5µM, you would need blood levels of 437.5ng/mL.

And that’s a problem, because above 80 or so ng/mL, ivermectin is pretty toxic.[3] To reach the lowest conceivable limit of inhibitory action on 3CLPro would require doses that are incompatible with… well, life.

The alleged ‘Pfizermectin’ achieves this efficacy at a nanomolar level – at a concentration roughly ten thousand times lower.[4] At that concentration, Paxlovid is still well within its margin of safety.

This is not by accident, by the way. Paxlovid is the product of rational drug design, using computer models to identify a small molecule drug that is not only effective against an enzyme target but also has favourable pharmacokinetics. Far from merely ‘repackaging’ ivermectin, as the Pfizermectin theory alleges, this is an entirely new drug. It is structurally novel, and bears little resemblance, if any, to ivermectin. It was designed from the ground up to be orally effective, and be so at doses that are safe.


In the end, action happens in vivo — in real, living beings. Ivermectin might be a workable drug in vitro, where the model does not have to account for the inconvenience of actually keeping the host organism alive and breathing. To many of us who have looked at the maths of it, it has been clear that with the inhibitory concentrations required by ivermectin, it is never going to transition from the bench to bedside. Ivermectin does have a mechanism that counteracts SARS-CoV-2, the virus responsible for COVID-19. It’s just that it is way too weak to be of any practical use.

Pfizer’s drug developers have solved this through molecular dynamics simulations and rational drug design (and, in fact, have released a paper on just how they did it, which is very, very cool)[5]. Paxlovid is the result of this development — and it is far from being ‘Pfizermectin’ or repackaged ivermectin.

Until now, much of the activity has been focusing on debunking misconceptions about the COVID-19 vaccine and sensationalistic tripe about the pandemic. As curative non-vaccine treatments are entering the market, it will be important to provide the public with the proof they need to dispel dishonest notions about the “Covid pill”. First and foremost among these is that the “Covid pill”, whether molnipiravir or Pfizer’s Paxlovid, is not a repackaging of ivermectin in any conceivable sense.

And the strongest proof of that is the simplest, too: it works, in ways ivermectin is mathematically incapable of working.


1 Popp M, Stegemann M, Metzendorf MI, et al. Ivermectin for preventing and treating COVID-19Cochrane Database Syst Rev. 2021;7(7):CD015017. Published 2021 Jul 28. doi:10.1002/14651858.CD015017.pub2
2 Mody V, Ho J, Wills S. et al. Identification of 3-chymotrypsin like protease (3CLPro) inhibitors as potential anti-SARS-CoV-2 agentsCommun Biol 4, 93 (2021
3 González Canga, A, Sahagún Prieto, AM, Diez Liébana, MJ et al. The Pharmacokinetics and Interactions of Ivermectin in Humans—A Mini-review. AAPS J 10, 42–46 (2008).
4 Owen DR, Allerton CMN, Anderson AS et al. An oral SARS-CoV-2 Mpro inhibitor clinical candidate for the treatment of COVID-19. Science. 2021;eabl4784.
5 Zhang S, Krumberger M, Morris MA, Parrocha CMT, Kreutzer AG, Nowick JS. Structure-based drug design of an inhibitor of the SARS-CoV-2 (COVID-19) main protease using free software: A tutorial for students and scientistsEur J Med Chem. 2021;218:113390. doi:10.1016/j.ejmech.2021.113390
Chris von Csefalvay
Chris von Csefalvay is a data scientist/computational epidemiologist specialising in the dynamics of infectious disease and AI-based, data-driven solutions to public health problems.

You may also like

1 Comment

  1. Also see: if Paxlovid was really ivermectin in sheep’s clothing, we wouldn’t be looking at months of factory time to produce it.

Leave a Reply