Humoral immunity to infectious diseases is important for protection against foreign attacks and clinical disease. Quantification of the antibody response is often measured through serological assays, and from this, seroprotection and seroconversion are determined by examining the changes in antibody titres in individuals over time. However, this approach lacks insight into the mechanisms underlying antibody kinetics and does not have the predictive power to infer how different conditions can alter antibody titres, leaving a gap in our understanding of factors that can influence antibody responses.
Mathematical models provide a powerful tool to simulate and predict antibody kinetics to better understand the dynamics of antibody production, decay, and longevity of the response. In this talk, I present the use of an established antibody kinetics model (White et al., 2014) to assess the longevity of the antibody response and serological exposure markers of protection against SARS-CoV-2 and Plasmodium vivax malaria. Using comprehensive clinical data, I have estimated the half-life of the IgG response to key COVID antigens by accounting for the biphasic decay of both antibodies and plasma cells and used it to infer time of infection or vaccination. I have also applied the model to data collected from P. vivax infected individuals in Thailand to determine optimal antigens for a new screening approach (Longley et al., 2020). Additionally, I have inferred the unknown time of primary infection for individuals in the Thai data to determine whether they were likely to be carrying dormant parasites in their liver at the time of the study, which cannot currently be clinically diagnosed.
Overall, this work will emphasise how critical mathematical modelling is in understanding the duration of protection given by the antibody response and the mechanisms behind antibody production and decay. This is important for vaccine development in helping inform the selection of antigens and the optimal timing to give booster doses, and treatment strategies from estimating the time of infection of an individual.