Your original statement that there were few documented reinfections. They found 30 of them here under controlled study conditions, and prior infection provided less than 50% protection against B.1.351.
To my knowledge, there hasn't been a controlled study like this in Brazil. The in vitro results you cited suggest that while the immune escape of P.1 is not as great as B.1.351, it still is very substantial, and reinfection will occur in a substantial number of patients.
I disagree with your assessment of P.1 immune escape being “very substantial” based on those in vitro results. It’s on par with the UK variant, which has no evidence - real world or otherwise - or reputation at all for widespread reinfection or vaccine escape. Never mind that in vitro results don’t even tell the whole story of an in vivo immune response. To be perfectly honest, I’m struggling to see how you’re making that claim.
[1] https://www.washingtonpost.com/business/can-you-get-covid-tw...
And actually, Novavax later announced that actually, natural immunity does provide protection against the SA variant [1]:
> A previously reported initial analysis from the study through 60 days indicated that prior infection with the original COVID-19 strain might not completely protect against subsequent infection by the variant predominantly circulating in South Africa. However, the complete analysis of the South Africa trial indicates that there may be a late protective effect of prior exposure with the original COVID-19 strain. In placebo recipients, at 90 days the illness rate was 7.9% in baseline seronegative individuals, with a rate of 4.4% in baseline seropositive participants.
Moreover: the SA variant and P.1 aren't necessarily similar in terms of immunity evasion. For instance: SA isn't using the AZ vaccine at all because of how ineffective it is against their variant, but it works just fine against P.1. Or look at this paper [2]:
> Neutralization titres against P.1 were similar to those against B.1.1.7 and only a minority of samples failed to reach 100% neutralization at 1:20 dilution of serum, considerably better than neutralization of B.1.351, where titres were reduced 7.6-fold and 9-fold for the BNT162b2 Pfizer and ChAdOx1 nCoV-19 AstraZeneca vaccines respectively.
[1] https://ir.novavax.com/news-releases/news-release-details/no...
[2] https://www.biorxiv.org/content/10.1101/2021.03.12.435194v1