Read on Twitter
1. I suppose because I'm an evolutionary biology who has worked on emerging infectious diseases for 20 years, a number of people are asking me whether we should expect the #SARSCoV2 #coronavirus to mutate to become more dangerous—or more mild—as the pandemic progresses.
2. In principle, the virus could mutate in any number of ways, but two of the most important to think about involve
VIRULENCE: the amount of harm the disease causes to its host, including mortality risk,
and
TRANSMISSIBILITY: the ease with which new infections are generated.
VIRULENCE: the amount of harm the disease causes to its host, including mortality risk,
and
TRANSMISSIBILITY: the ease with which new infections are generated.
3. Let's start with transmissibility. Natural selection will favor variants of the virus that are more likely to transmit and initiate new infections. For example, a variant might reproduce better in the upper respiratory tract and thus be spread more in respiratory droplets.
4. If such variants arise, we would expect them to spread, because they would produce more new infections than the previous "ancestral" strains.
We certainly can't rule out the possibility of this happening, but I don't there is good evidence that it has happened yet.
We certainly can't rule out the possibility of this happening, but I don't there is good evidence that it has happened yet.
5. It's hard to predict for certain, but my personal opinion is that given the rates at which we seeing mutations in the virus and the structure of its genome, there is not strong reason to expect more transmissible variants to evolve and become widespread over the next year.
6. Now let's consider virulence.
One argument has it that diseases should evolve not to kill their hosts too quickly, because a dead host doesn't transmit to other people. There are a lot of problems with this argument both in general and for #COVID19 in particular.
One argument has it that diseases should evolve not to kill their hosts too quickly, because a dead host doesn't transmit to other people. There are a lot of problems with this argument both in general and for #COVID19 in particular.
7. People used to take this reasoning to the conclusion that microbial pathogens should co-evolve with their hosts to cause minimal harm. The problem with that argument is that there tend to be *virulence-transmissibility tradeoffs*.
8. To transmit effectively, a pathogen generally needs to make a lot of copies of itself within the host. Doing this tends to cause harm to the host, either directly or as a byproduct of the host's immune response.
Of course there are exceptions. Biology always has exceptions.
Of course there are exceptions. Biology always has exceptions.
9. So we might expect a pathogen to evolve intermediate virulence. Paul Ewald has proposed this as a reason why we shouldn't worry too much about H5N1 bird flu. It may now kill over 60% of people infected, but if it takes off in humans shouldn't it evolve to become less lethal?
10. I've been very critical of this line of thinking:
Perhaps in the long run, a pandemic H5N1 would decline in virulence. But to misuse John Maynard Keynes, in the long run we are all dead. Especially in a bird flu pandemic.
Beneficent evolution would be little consolation.
Perhaps in the long run, a pandemic H5N1 would decline in virulence. But to misuse John Maynard Keynes, in the long run we are all dead. Especially in a bird flu pandemic.
Beneficent evolution would be little consolation.
11. More generally, predicting virulence evolution is hard, especially for an emerging disease. When a new pathogen enters the human population, it is unlikely to be adapted to maximize transmission. It might be too virulent, or too mild.
12. Basically, any old thing could happen, depending on what mutations arise first. In this short piece ( http://blogs.nature.com/journalclub/2008/05/carl_bergstrom.html), I briefly describe a lovely modeling paper to this effect by Jim Bull and Dieter Ebert ( https://onlinelibrary.wiley.com/doi/full/10.1111/j.1752-4571.2007.00003.x).
13. If any old thing could happen, then couldn't #COVID19 evolve to get worse—more virulent—as the pandemic progresses?
Sure, it *could*. I'm not tremendously worried about it. In part, we aren't seeing signatures of this kind of evolution thus far.
Sure, it *could*. I'm not tremendously worried about it. In part, we aren't seeing signatures of this kind of evolution thus far.
14. But there's another really important reason. At present, a *very* conservative estimate would have 200,000 active cases (basically, all the known ones and no others).
Suppose a new mutation arises today that changes the infection fatality rate from around 1% to around 10%.
Suppose a new mutation arises today that changes the infection fatality rate from around 1% to around 10%.
15. Sounds dreadful. But what are you chances of getting this more deadly strain if you catch #COVID19 tomorrow?
About 1 in 200,000.
About 1 in 200,000.
16. And here's the thing. Unless the new more virulent strain has a huge transmission advantage, your chance of catching it instead of regular COVID *stays* at about 1 in 200,000 going forward into the future, because the new strain stays rare relative to the old one.
17. So now once this happens, you've got a 1/200,000 chance of a 10% death rate and 199,999/200,000 chance of a 1% death rate when you catch COVID19. In other words, about 1%, just like before. I wrote about this sort of thing w/ @mtanaka000 & @BMAlthouse: https://www.ncbi.nlm.nih.gov/pubmed/25376480
18. In short, it's possible that #COVID19 will mutate in various ways. Predicting these one-time events is extremely difficult.
But I'm not expecting evolution to grant us a reprieve—or to make everything much worse.
Mostly likely, what we see is what we get.
/fin
But I'm not expecting evolution to grant us a reprieve—or to make everything much worse.
Mostly likely, what we see is what we get.
/fin
Addendum: Trevor Bedford's latest thread provides much needed context about strains vs. mutations, evading immunity, differences from influenza, etc. https://twitter.com/trvrb/status/1242628550563250176
