Mathematical Modeling of Vaccines That Prevent SARS-CoV-2 Transmission

SARS-CoV-2 vaccine clinical trials assess efficacy against disease (VE_DIS), the ability to block symptomatic COVID-19. They only partially discriminate whether VE_DIS is mediated by preventing infection completely, which is defined as detection of virus in the airways (VE_SUSC), or by preventing symptoms despite infection (VE_SYMP). Vaccine efficacy against transmissibility given infection (VE_INF), the decrease in secondary transmissions from infected vaccine recipients, is also not measured. Using mathematical modeling of data from King County Washington, we demonstrate that if the Moderna (mRNA-1273QS) and Pfizer-BioNTech (BNT162b2) vaccines, which demonstrated VE_DIS > 90% in clinical trials, mediate VE_DIS by VE_SUSC, then a limited fourth epidemic wave of infections with the highly infectious B.1.1.7 variant would have been predicted in spring 2021 assuming rapid vaccine roll out. If high VE_DIS is explained by VE_SYMP, then high VE_INF would have also been necessary to limit the extent of this fourth wave. Vaccines which completely protect against infection or secondary transmission also substantially lower the number of people who must be vaccinated before the herd immunity threshold is reached. The limited extent of the fourth wave suggests that the vaccines have either high VE_SUSC or both high VE_SYMP and high VE_INF against B.1.1.7. Finally, using a separate intra-host mathematical model of viral kinetics, we demonstrate that a 0.6 log vaccine-mediated reduction in average peak viral load might be sufficient to achieve 50% VE_INF, which suggests that human challenge studies with a relatively low number of infected participants could be employed to estimate all three vaccine efficacy metrics.