Mapping immune imprinting zones enables predictive vaccination optimization

Immune imprinting, where prior exposures shape antibody responses to variant antigens, remains a central obstacle to optimizing vaccination against evolving viruses. Here, we present DynaVac, a mechanistic framework that models antigen-specific B cell dynamics, particularly the competition between memory and naive compartments across antigenic distances. Calibrated on neutralization titers from murine and human studies spanning diverse SARS-CoV-2 vaccine platforms, DynaVac accurately predicts antibody responses across complex heterologous and multivalent regimens. In silico simulations reveal three imprinting zones—protection, pitfall, and breakthrough—that determine when updated vaccinations amplify, suppress, or bypass preexisting immunity. Unlike prior models limited to qualitative or single-exposure settings, DynaVac integrates empirical cross-neutralization matrices and enables prospective simulation of continuous booster responses across antigenic variants, dosages, and intervals. DynaVac also provides an actionable strategy for guiding real-time vaccine updates and strain selection. While DynaVac is calibrated on SARS-CoV-2, its structure is generalizable to other fast-evolving pathogens.