Informing risk stratification and vaccine policy via a predictive mathematical model of SARS-COV-2 infection

Source abstract

Abstract SARS-CoV-2 human challenge studies enabled unprecedented high frequency temporal measurements of immune determinants of viral control. We integrated prior knowledge of SARS-CoV-2 replication, innate, cytotoxic T cell and antibody responses into a dynamic model which captured viral-load kinetics in seronegative volunteers challenged with wildtype virus. The model predicted measured innate and adaptive immune dynamics. Perturbation analyses indicated that innate immunity limits peak viral load, whereas antibodies principally mediate clearance while cytotoxic T cells become critical in antibody deficiency. Pre-existing antibodies suppressed both viral replication and subsequent immune activation- a prediction validated in a second challenge study of seropositive participants infected with the SARS-CoV-2 Delta variant. The model explains how specific immunodeficiencies may increase susceptibility to severe or persistent infection, supporting clinical risk stratification. It suggests that T cell-targeted vaccination strategies may provide limited protection without effective neutralising antibodies, and that attenuated boosting after low-dose exposure requires repeated vaccination to reduce breakthrough infections, informing public health policy. One Sentence Summary A mathematical model of SARS-COV-2 infection predicts innate and adaptive immunity, and informs risk stratification and vaccine policy.