Capturing SARS-CoV-2 immune landscapes to inform future strategies for COVID-19 vaccination in a high-income setting: a mathematical modelling study

In an era of endemic SARS-CoV-2 transmission, countries are continuing to evaluate how best to schedule ongoing COVID-19 booster vaccinations. Mathematical modelling provides a useful tool to predict the benefit of future vaccination strategies, incorporating the loss of protection due to waning immunity and strain mutation. We adapted a combined immunological-population transmission model for SARS-CoV-2, to better capture contemporary understanding of exposure- and vaccine-derived immunity, to simulate ongoing endemic transmission of SARS-CoV-2 in a highly exposed high-income setting. We used this model to estimate the impact of targeted booster dose strategies in the older population, both in the context of continued circulation of the current dominant viral strain, and in the presence of a new antigenically distinct variant. We found that at the population level, an annual COVID-19 vaccine booster dose to the 65+ years population at 60% coverage could avert 8–16% of hospitalisations over a single wave, depending on how well-matched the vaccine is to the circulating SARS-CoV-2 strain. With lower coverage of 40%, estimated median impact was between 6% and 11%. A second booster dose to the 75+ population after 6 months was particularly beneficial if a new distinct variant strain increases the magnitude of the wave. Of the vaccine scenarios explored, we found that increasing uptake of the annual booster dose in the 65+ population is likely to have a larger impact on hospitalisations than optimising dose timing. This adapted model captures endemic viral transmission and could readily be used to explore vaccine impact across other settings.