Background
High rates of vaccination and natural infection drive immunity and redirect selective viral adaptation. Updated boosters are installed to cope with drifted viruses, yet data on adaptive evolution under increasing immune pressure in a real-world situation are lacking.
Methods
Cross-sectional study to characterise SARS-CoV-2 mutational dynamics and selective adaptation over >1 year in relation to vaccine status, viral phylogenetics, and associated clinical and demographic variables. Findings: The study of >5400 SARS-CoV-2 infections between July 2021 and August 2022 in metropolitan New York portrayed the evolutionary transition from Delta to Omicron BA.1-BA.5 variants. Booster vaccinations were implemented during the Delta wave, yet booster breakthrough infections and SARS-CoV-2 re-infections were almost exclusive to Omicron. In adjusted logistic regression analyses, BA.1, BA.2, and BA.5 had a significant growth advantage over co-occurring lineages in the boosted population, unlike BA.2.12.1 or BA.4. Selection pressure by booster shots translated into diffuse adaptive evolution in Delta spike, contrasting with strong, receptor-binding motif-focused adaptive evolution in BA.2-BA.5 spike (Fisher Exact tests; non-synonymous/synonymous mutation rates per site). Convergent evolution has become common in Omicron, engaging spike positions crucial for immune escape, receptor binding, or cleavage. Interpretation: Booster shots are required to cope with gaps in immunity. Their discriminative immune pressure contributes to their effectiveness but also requires monitoring of selective viral adaptation processes. Omicron BA.2 and BA.5 had a selective advantage under booster vaccination pressure, contributing to the evolution of BA.2 and BA.5 sublineages and recombinant forms that predominate in 2023. Funding: The study was supported by NYU institutional funds and partly by the Cancer Center Support Grant P30CA016087 at the Laura and Isaac Perlmutter Cancer Center.