Androgen Receptors for COVID-19

 By Derek Lowe

There’s a report of an interesting small-molecule drug effort against the coronavirus that seems to have produced rather significant results. The idea goes back to effects that were noticed last year – for example, in this population-based study from Italy. It’s been known since the early days of the pandemic that males were overall more susceptible to severe disease than females, and there have been a number of hypotheses put forward to try to explain this. But an answer might lie back in the virus’s mechanism of infection. As the world knows, the current coronavirus uses the ACE2 protein as a cellular entry point, followed by a key protein cleavage effected by the nearby TMPRSS2 transmembrane serine protease. That one’s involved in more than one viral infection route, and the idea of using serine protease inhibitors as antivirals had already been tried against type A influenza  – but, it has to be said, without any dramatic success. Favipiravir is one such molecule, but it hasn’t made much of a dent in the pandemic, either.

But there are other ways to target this mechanism. It turns out that TMPRSS2 expression is linked to the transcriptional-activating activity of the androgen receptor. The Italian work linked above found that men who were taking androgen receptor antagonists (generally as a treatment for prostate cancer) seemed to be at significantly lower COVID-19 risk, which result is especially striking considering that pre-existing cancer in general is a risk factor for severe coronavirus outcomes. Instead of trying to inhibit TMPRSS2 activity at its active site, AR antagonists keep it from being expressed in the cell membrane in the first place.

This line of thought has led to several followups. Here’s a proposal from last June that androgen antagonists be tried out in this fashion, and here’s a later study from Michigan that established that specific cell types in the airway do indeed co-express androgen receptors and TMPRSS2, which is then regulated in the lung by AR activity. In July, a Chinese company (Kintor Pharmaceutical) announced that it was providing an investigational androgen receptor antagonist (proxalutamide) for a clinical trial to be run in Brazil. And that trial (involving 588 patients) has now read out. The press release says that there were very significant reductions in disease severity, in overall mortality, and in the length of hospitalization. These look like strong results, and I’m glad to see them. Anti-androgen therapy is pretty strong stuff, so it’s good that the course of treatment is reasonably short. This is indeed the receptor through which testosterone exerts most of its effects, so the effects of long-term treatment in men can be quite noticeable (women have fewer side effects, but non-zero). But blockade of the androgen receptor in this way is generally a cleaner route than (say) blocking the synthesis of testosterone itself, because there are quite a few other pathways involved with testosterone metabolites, etc.

Now, one thing to note is that proxalutamide itself is still an investigational drug – to the best of my knowledge, it hasn’t been approved anywhere. Here’s a trial in Florida looking at a much more common AR antagonist (bicalutamide), but unfortunately it’s only 100 patients. And here’s one looking at the combination of bicalutamide and the serine protease inhibitor camostat at Johns Hopkins, but it’s only 60 patients. I think we might be in a more useful place if the Brazil trial had been run with an already-approved AR antagonist, but this is one of those situations when interests and feasibility don’t overlap as well as they should. We have in general had far too many small, underpowered trials of possible therapies during this pandemic (a topic that will get a blog post all its own!) I hope that if the effect size of this treatment is as large as it appeared in the proxalutamide trial that these others will show something as well, but we’ll see.

Back on the biological level, you may be wondering how the expression of some transmembrane serine protease gene ended up being tied to the testosterone receptor in the first place. Well, that’s gene expression, for better or for worse. It’s a tangle – all sorts of things are wired crossways to all sorts of other things, in patterns that sometimes seem reasonable and sometimes seem random. Remember, there are far more genes out there than there are transcription factors, so everything at that level is doing multiple duty and affecting the expression of a long list of things. As always, evolution’s one question is “Did you survive to produce offspring” and its one rule is “Whatever works”. It’s important to remember that the AR isn’t just a male thing: women have androgen receptors as well, and those pathways are most certainly not silent. For a recent example, here’s evidence that for estrogen-receptor-positive breast cancer, you might also want to activate the androgen receptors as well in therapy. This hypothesis has been argued about for some time, but evidence for it has been increasing, and clinical trials around it are underway.

Let’s hope that the AR blockade idea can have a positive impact in the current pandemic, then. And we can also hold this idea if we face another virus that uses the ACE2 receptor for cell entry. That was the case with SARS in 2003, and we might see it again!

https://blogs.sciencemag.org/pipeline/archives/2021/03/11/androgen-receptors-for-covid-19