Speaker
Description
Understanding how germinal center (GC) dynamics generate antibody responses against emerging viral variants is important for vaccine modeling. Here we develop a stochastic GC model formulated in an antigenic space framework to study antibody responses to SARS-CoV-2 vaccination. Naïve B cells and antigens are represented as coordinates in antigenic space, and affinity is determined through a bit-matching score that drives B cell selection and differentiation into plasma and memory cells. Plasma cells (PCs) are classified as aligned or non-aligned with the target variant, allowing the model to capture both variant-specific and cross-reactive antibody production.
The model parameters were calibrated using experimental data (50% neutralizing titer) across multiple variants from different vaccination regimens. The simulations reproduced the qualitative antibody dynamics observed in the data. The model suggests that early antibody production is primarily driven by PCs targeting the vaccine-matched variant, whereas broader cross-variant responses emerge later and are sustained by longer-lived PC populations generated during germinal center maturation and booster-induced reactions. Memory B cells contribute to improved affinity following booster vaccination. This framework provides a mechanistic approach to linking GC selection dynamics with antigenic distance, which could help to interpret scenarios involving vaccination and breakthrough infection with emerging variants.
