[variaga]

The rough theory - as I understand it - is that the virus is composed of other things as well as the spike protein e.g. the proteins in the viral shell. The immune system can learn to attack _whatever_ foreign materials it encounters. So a person that clears the infection might have a strong immune response to the spike protein, but the response could also be focused on some other part of the virus (or a mixture of different viral features).

Problem is, the "some other part" can mutate more easily into a form that the immune system no longer recognizes than the spike protein can, because the spike protein has to keep its form in order to react with the ACE2 receptors and enter a human cell. So the concern isn't that a real infection doesn't leave you immune to the virus you got sick with, but that the immunity might not apply or be as robust against a variant where the "some other part" has mutated into a different form.

The mRNA vaccines only cause your cells to make the (difficult to significantly mutate) spike protein, so the expectation is that vaccinated people would have a robust immune response against any variants which have that spike protein, regardless of whatever else changed, and a variant with an unrecognizable mutation of the spike protein would be less infectious anyway, as it could no longer successfully enter human cells either.

The results of the paper seem to indicate that this is not a problem in practice (so far - new variants are still evolving), but it's nice that someone checked.

[starfallg]

A 2-dose vaccination regime (separated by 8-12 weeks) also gives our immune system time to build up the spike protein. An actual infection doesn't last as long and hence our immune system doesn't process it in the same way.

[josephcsible]

> separated by 8-12 weeks

Who's doing that? Isn't the separation 3 weeks for Pfizer and 4 for Moderna?

[mlyle]

Canada/the UK have been doing that.

Odds are it does make things more effective rather than less, based on our experience with other vaccines. But IMO it's not wise, because it wasn't studied.

Still, 2 huge peaks separated by 3-4 weeks provokes really big immune responses.

[BeefWellington]

> Odds are it does make things more effective rather than less, based on our experience with other vaccines. But IMO it's not wise, because it wasn't studied.

It has since been studied[1] (preprint) and that was exactly what was found -- delaying the second dose provoked a stronger response, similar to many other vaccines.

[1] https://www.birmingham.ac.uk/news/latest/2021/05/covid-pfize...

[mlyle]

Sure, so far preliminary data looks OK. I still don't think it was a wonderful roll of the dice-- it didn't lower susceptibility so much in the short term (because a single dose's efficacy is limited, especially in the old) and it could have made things much worse.

[ufo]

A key limitation of the study is that it's in the USA, where some of the most concerning variants are not as widespread.

For example, in Brazil there is evidence that the P.1 variant can lead to reinfections.