Hacker News Reader

Comment by ohazi | Hacker News Reader

> How long immunity lasts for following covid-19 infection is the biggest unknown. Comparison with other Coronaviridae suggests it may be relatively short-lived (i.e. months).

Anyone have sources for this?

I've been skeptical of the rumors about reinfection, because they seem to be based on tests coming back positive after recovery, but the tests are rt-PCR based, and your body will be shedding leftover bits of RNA as it flushes out viral remnants long after recovery, so this seems expected.

As far as I'm aware, immunity to most viruses is long-lasting or permanent, with the obvious exception being if the virus mutates enough to look different to your immune system. Is this assumption wildly off?

View on HN

I remember reading they expect immunity to last as little as 6 months for Covid-19, found this paper about the SARS outbreak of 2002 which was also a type of corona virus, saying patients had an average immunity of 2 years.

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2851497/

Wikipedia-grade non-biologist here, so again, feel free to correct me if I'm wrong.

As far as I'm aware, antibodies don't stick around after they're no longer needed... They're produced by B cells when an infection is detected, and after an initial infection, some B cells that learned how to make effective antibodies stick around in your body as "memory B cells" - these will produce the same antibodies again upon reinfection.

Counting antibodies a few years after a SARS infection doesn't seem like it would tell you much about immunity.

It's correct that the memory cells in your body will last a long time. However, (I think) what the parent is getting at is that within a relatively short amount of time the virus will have mutated just enough to evade detection by those memory cells.

The other four coronavirus' that form part of what we call the common cold all mutate readily and come back and reinfect us year after year. This one could be different but there is plenty of good historical reason to believe it will be like the others.

Just to note if you are going to use SARS as a comparison:

SARS-CoV-1 completely disappeared after two flu seasons and hasn't been seen in decades now.

It's probably accurate. But the assumption that this is a game changer is more questionable.

This variant of CV is fairly infectious and makes some people very ill. It also kills a non-trivial percentage.

So the worst case is an incredibly dangerous virus that mutates every season. We'd have to get used to taking extreme measures, or losing around 1% of the population every year.

But the usual trend is for viruses to mutate in a less dangerous direction. And immunity isn't binary. It's more like molecular pattern matching with variable memory. So the population as a whole wouldn't be starting from scratch every season - any more than it does with colds or flu, even though they both mutate quickly.

Short term we're likely to have the equivalent of a few years of very bad flu seasons, and longer term that would settle down to a less dangerous cycle - still lethal for some people, but not a perpetual state of danger and crisis.

> We'd have to get used to taking extreme measures

Or develop drugs to treat the people who become very sick (up to bilateral interstitial pneumonia) to avoid or reduce hospitalization. On the medium term this looks like a possibility (vaccines, if they work, would help long term). If that happens, I believe your scenario of a "less dangerous cycle" will likely materialize earlier. People would still get sick, but at least we'd be able to treat them.

> But the usual trend is for viruses to mutate in a less dangerous direction.

This trend is for viruses that are in danger of killing their hosts before spreading.

This coronavirus is so infectious that it's nowhere near that. Hence it can even mutate into more lethal form without risking its ability to spread.

Mutations are random. What is not random is how human society selects from the pool of random mutations.

I am thankful that we do not have a global ground war in progress right now. I believe that political pressure to keep the healthy isolated on the front lines, with the sick going home to society -- this reverse version of containment -- was 1918 society selecting the more lethal mutations for increased reproduction.

From this standpoint, the herd immunity and flatten the curve approaches are no different. Both put severe versions of the mutated virus in an a hospital, which is an environment where eventually most hosts will have the virus, effectively deselecting virulence.

That said, not overwhelming hospitals probably means having more healthy doctors available for the virus to infect-- a slight selective pressure for virulence. This would be in contrast to herd immunity -- letting both a massive spike of people die along with doctors and then turning hospitals into simple pre-death, low-R_0 isolation wards.

The issue there though is that overwhelming hospitals via the herd immunity approach would decrease our ability to learn about the virus scientifically. Our ability to learn about pathogens and modify our behavior accordingly is probably our best hope for survival as a species, I think, so perhaps we should not overwhelm the hospitals with the herd immunity approach.

> perhaps we should not overwhelm the hospitals with the herd immunity approach.

Which is great because that's the same thing everybody selfishly would like to do just to have access to healthcare when he needs it.

Everybody in the world seems to understand that except perhaps Boris Johnson.

This virus is so bad that it is making whole countries take quarantine measures. That's probably not optimal for spreading itself.

Evolution doesn't care directly about spreading. Just about survival and this virus is currently at no danger of being eradicated by measures in effect.

More spreading means more survival. A virus that we could reasonably live with would have more surviving copies than this one.

If you can infect all hosts you can't do better. And I think this virus can, even being twice as lethal as it is.

I just read this preprint: https://www.medrxiv.org/content/10.1101/2020.03.05.20031773v...

People over ~50 have a ~50% chance of not showing symptoms, under 50 is ~10%. I wonder if there was a flu/cold epidemic 50 years ago that is causing the difference?

I know that some flus have different effects depending on age group due to cohort exposure decades ago.

Though it is a preprint, so could be all sorts of sampling reasons for that outcome.

The https://en.wikipedia.org/wiki/1968_flu_pandemic perhaps?

https://www.biorxiv.org/content/10.1101/2020.03.13.990226v1

So they tried to reinfect monkeys 28 days after the initial infection, and the monkeys were immune. This looks good, at least for the very short term. For the long term, see previous link

Reply on news.ycombinator.com