Hacker News Reader

Comment by swader999 | Hacker News Reader

In general, respiratory viruses tend to get less severe over time. With this mass vaccination approach it may not turn out that way by some accounts I've read.

View on HN

In the book “spillover”, notably written before Covid (so quite prophetic read from today’s perspective), the author discussed this claim. Although it often happens, it is by no means guaranteed (at least on satisfactory timelines), and there are counter examples.

One is the evolution of myxomatosis in Australia, it is a disease affecting rabbits. The disease split into 4 strains. Eventually the dominant one become, IIRC, one that had the slowest progression, but also was the deadliest overall.

Buns infected with that strain had larger chances of passing it on (since they were alive longer), and yet they would be more likely to die.

I guess evolution doesn’t care, and if a strain is more deadly, yet more persistent, it will win.

The easier way to look at this is that from the perspective of evolutionary pressure what happens to a host (or parents, for that matter) once reproduction has taken place is irrelevant. So there is no selection pressure for mortality, that's just another outcome, unless it happens too quickly in case the virus has less chance to make it to the next generation.

Reductio ad absurdum: if a virus would kill on first contact there would never be any time for it to spread, but once a virus has spread the host is not all that interesting unless it could be caused to continue to spread. So whether the host lives or dies after that won't cause that particular virus to be more or less successful.

I would go further, and say we have no evidence that any human virus has evolved to become less virulent as it becomes 'endemic'. Instead, susceptible individuals simply die off (e.g. smallpox in the Americas).

In order for there to be an evolutionary pressure toward becoming less severe, having higher severity has to give some disadvantage - e.g. killing the host before it can spread the disease or giving stronger symptoms so hosts can be isolated before they can spread it. The corona virus is pretty unique in how much it can spread before/without any symptoms showing up, which should mean there's relatively less room for improvement by lowering the severity.

I think it is the other way around, there is no pressure to maintain a high severity as long as the virus can spread thus leading to mutations that lose the severe traits.

Viruses aren't trying to kill their hosts, they're trying to replicate. But to replicate they kill the cells they're in.

COVID's severity is because it's making a tradeoff between the time it takes the immune system to destroy it, vs the need to get a host walking around and socializing while breathing it out onto new hosts. The lethality is a side-effect of its replication strategy.

I thought the lethality was just the immune system running havoc. If a virus could achieve it, it would remain symptom-less and just spread.

Viruses lyse (explode) the cells they take over in order to exit the cell, since packaging for membrane transport is much more complicated.

Unchecked, they would just exponentially destroy your tissue.

Cytokine storm responses are a particular kind of immune system over-reaction that can make relatively benign viruses much more dangerous to a healthy host then a more immunologically suppressed host, but that's not the behaviour of COVID-19 which is showing pretty standard lethality profiles.

Surely such mutations can and will happen. But will they become dominant, if they do not provide any advantage?

sometimes the "best" team doesn't win the league. Its plausible.

Perhaps if it requires less energy to replicate and/or it invokes a weaker immune response?

Uh, can you point at an example of a pressure to kill your host?

No that's why I said there is no pressure to be severe. Killing hosts doesn't provide many benefits (unless you spread via people touching dead bodies such as Ebola).

Asymptomatic spread is a myth - https://jamanetwork.com/journals/jamanetworkopen/fullarticle...

The study you cite doesn't seem to say what you want it to say.

> Controlling for differences across studies, secondary attack rates were higher in households from symptomatic index cases than asymptomatic index cases, to adult contacts than to child contacts, to spouses than to other family contacts, and in households with 1 contact than households with 3 or more contacts.

It acknowledges asymptomatic spread.

Can someone explain whether corona is unique in this attribute of spreading versus other viruses? I would expect most cold-related viruses or other viruses in the human virome to exhibit this pattern.

As you suspect, Coronavirus isn't unique in having an incubation period, or in being infectious during the latter part of the incubation period.

Of course, there's questions of degree.

Citations? I think I've heard that too, but if you've read something recently that would help me separate a study from a rumor.

The argument one frequently reads/hears on this is, that a virus has no evolutionary advantage to kill its host and if the virus becomes fitter in terms of infecting new hosts, they can neglect functions in their genome that lead to deadliness.

A respiratory virus may have an evolutionary advantage if it doesn't hit the organism so hard that the host stays home and isolates. Mild symptoms may increase the likelihood for socialising for example, thus there could be evolutionary pressure for a milder form to develop.

I would assume that these are just general observations and it won't allow a clear prediction where COVID19 variants are headed. But there are theories that other coronaviruses have been more aggressive initially until they became the milder forms that are nowadays endemic.

Of course we know plenty viruses that have evolved and are still deadly, so this isn't something I would bet on.

Host availability is likely the greatest pressure I would think. Then severity or its lack, given our numbers, would be less a driver of genetic change than would random mutations.

https://en.wikipedia.org/wiki/Marek%27s_disease#Prevention is the canonical example:

"However, the leaky vaccine changes this evolutionary pressure and permits the evolution of highly virulent strains. The vaccine's inability to prevent infection and transmission allows the spread of highly virulent strains among vaccinated chickens. The fitness of the more virulent strains is increased by the vaccine."

See:

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4516275/

Andrew Read on the issue:

https://theconversation.com/vaccines-could-affect-how-the-co...

It seems kind of unlikely that the scientific world would have overlooked this concept

Isn't Delta vs Alpha a pretty convincing counter argument? It was more transmissible, more severe and better at avoiding acquired immunity.

Source for the Delta variant being more severe? I hadn’t heard that.

Are we certain Delta was/is more severe? Since it’s more contagious, might it also be producing a lot more mild/asymptomatic cases that don’t get reported? Severity seems like a difficult thing to measure reliably if you can’t be sure about the denominator.

Multiple countries have sustained mass testing of at least certain groups of population. The argument about undetected asymptomatic cases would be valid in early 2020 when tests were scarce, but now there are good continuous metrics from people who get tested even if asymptomatic.

Gotcha—I’d be interested in links to some of this research. I wasn’t aware there was conclusive evidence that it’s more severe.

Oh, I'm not following the news on Delta or any other variants much, however, I just know people who are measuring the prevalence of variants in the general population in reasonable ways which would definitely cover also asymptomatic people (one is mass testing - e.g. right now 100% of local kids are getting weekly tests in schools, and all hospital admissions get tested even if it's e.g. a car crash, but there's also the viral analysis in sewer system, which is a cool way to get a total perspective on large populations), so studies about the severity of Delta (whatever they are saying) should not be distorted by the particular problem of asymptomatic cases not getting reported, the researchers now have good tools to get the "denominator" part correct.

Reply on news.ycombinator.com