Issue #66: Where are the small molecule drugs to treat COVID-19?

Written by Nobel Laureate Professor Peter Doherty

While we are all very aware of how quickly effective vaccines to prevent SARS-CoV-2 infection emerged, we’ve not been hearing a great deal about small molecule therapeutics directed specifically at this novel coronavirus. The public discussion so far has focused on repurposed drugs, ranging from the antimalarial Hydroxychloroquine, to Ivermectin – developed initially to treat heart worm in dogs and used very effectively to treat river blindness – to various ribonucleoside analogues. In each case, there has been some molecular rationale for going down that road, plus a lack of solid evidence re efficacy.

As anyone who has ever engaged with the process of drug evaluation and licensing would realise, testing a drug for therapeutic value and/or prophylaxis in COVID-19 requires properly constituted, double-blind trials (#56). The disease is just too variable in clinical presentation for any anecdotal evidence to be taken seriously and, particularly with Hydroxychloroquine, inadequate, underpowered trials drove confusion and misrepresentation. Unlike the original SARS virus of 2002, SARS-CoV-2 multiplies to the high titres that enable transmission early in the course of infection with, as is the case in influenza, any treatment with antivirals likely being most effective if given concurrent with initial presentation. Uniquely because of the unprecedented, widespread use of PCR testing for diagnosis, that could have been done with COVID-19. But, with a drug like Hydroxychloroquine that does have known toxicity, treating people with mild (or no) symptoms would not be a strategy with much appeal to medical professionals.

As a consequence, with drugs being given, initially at least, to patients with substantial symptoms, much of the trial work that was done may not have been especially useful as the therapy was being instituted too late in the course of the disease. That’s why some competent professionals persisted with testing Hydroxychloroquine as a possible early treatment. In the end, though, the general conclusion was that it did not have a useful part to play in therapy. The drug that looked more promising and was used extensively in hospitals – it isn’t easy to manufacture in bulk and supply was an issue – was the repurposed (from Ebola treatment) nucleoside analogue Remdesivir. Though it continues to be recommended for use in some clinical settings, it hasn’t proved to be the ‘magic bullet’ that everyone had been hoping for.

The value of antiviral small molecules versus vaccines is, of course, that the drugs target very conserved proteins (e.g. viral proteases) that are required for the replication of all coronaviruses. As a consequence, they can potentially deal effectively with mutant SARS-CoV-2 strains (or with any novel coronavirus) though, as with any chemotherapy, it is wise to give at least two different therapeutics targeting different pathways to ensure that functional mutants do not emerge (#6, #64, #65). Still, the mutation of a key viral protease that must retain enzyme function is even more likely to lead to a serious ‘fitness cost’ than is the case for some change to an antigenic site on a virus surface protein, in this case the SARS-CoV-2-spike.

Progressing through the repurposed antivirals testing pipeline is a another nucleoside analogue Molnupiravir (MK4482 developed for use against influenza) that works to reduce the replication of SARS-CoV-2 in the upper respiratory tract of ferrets, with a consequent diminution of transmission to naïve animals. Merck has already entered into voluntary license agreements with established Indian generics manufacturers, the companies that supplied AIDS drugs to the poorer nations. Molnupiravir has the advantage over Remdesivir in that it can be taken as a pill and does not have to be given intravenously. Even so, there’s a problem with all these nucleoside analogues. As they target the viral genome they are, of course, potential mutagens!

Then Pfizer has two antivirals in Phase 1 clinical trials: PF-07304814 has to be given intravenously while PF07321332 can be taken orally. Both target an enzyme (a protease) encoded by the SARS-CoV-2 genome that’s likely common to the CoV family and is essential for virus replication. In addition, our researchers at the Walter and Eliza Hall Institute and the Doherty Institute are, as must be the case in many laboratories across the planet, screening compound libraries and other likely chemicals for antiviral effects. Both Parkville-based efforts have identified possible targets for further development but, as we all understand, it is a very long path from candidate identification to testing, then eventually rolling out a useful drug.

That leaves the question of how any small molecule antiviral can optimally be used in COVID-19. Drugs will clearly be most effective if given early, but it would take very decisive actions before that would be likely to happen in a normal clinical or public health setting. However, especially as we consider how a fully (or 70-80%) vaccinated Australia can safely open up to a world where mutant SARS-CoV-2 strains are circulating, automatic treatment with an effective antiviral at time of positive PCR diagnosis is worth thinking through as a strategic possibility, even if it was available only to those who are highly vulnerable, such as the elderly or the immunosuppressed. Obviously, such treatments would work best if it just meant requiring people to ‘pop a pill or two’.

The above is an updated version of an invited editorial published in the Autumn 2021 issue of Pharmacy GRIT, the journal of the Society of Hospital Pharmacists of Australia