Immune imprinting and vaccine interval determine antibody responses to monovalent XBB.1.5 COVID-19 vaccination

Introduction

The World Health Organization (WHO) declared the end of COVID-19 as a public health emergency in May 2023¹, but SARS-CoV-2 variants continue to emerge and spread². To maintain population immunity, the XBB.1.5 monovalent vaccine was subsequently approved in 2023 and was shown to successfully induce cellular responses and neutralizing antibodies against circulating variants^{3, 4, 5–6}. The JN.1-adapted vaccines were further formulated to match prevalent strains in 2024⁷. However, the diversity and potency of new immune responses generated through updated COVID-19 vaccines may be influenced by immune imprinting following exposure to historic spike proteins during prior vaccination or natural infection⁸.

Immune imprinting has been extensively explored for influenza and is characterized by reduced induction of neutralizing antibodies against novel variants of a viral antigen upon exposure. Back-boosting is also highly characteristic of imprinting and describes boosting of titers against ancestral antigen upon exposure to a novel variant⁸. Recognition of immune imprinting in the context of COVID-19 initially arose with observations that infection resulted in back-boosting of binding antibodies against seasonal coronaviruses OC43 and HKU1⁹. Ancestral Wuhan-Hu-1 infection prior to Omicron BA.1 infection resulted in poorer Omicron-targeting antibody responses compared to infection with BA.1 alone¹⁰. Similarly, vaccination with monovalent Wuhan-Hu-1 prior to infection with Alpha or Delta variants generated weaker variant-specific responses than in unvaccinated, infected individuals¹¹. As the first widely administered variant-adapted vaccine, the bivalent Wuhan-Hu-1/BA.4-5 vaccine failed to significantly enhance Omicron-neutralizing antibodies compared to monovalent Wuhan-Hu-1 boosting, likely due to imprinting by the ancestral spike protein^{12, 13–14}. Consistent with this observation, minimal additional protection was provided against COVID-19 infection and hospitalization in bivalent versus Wuhan-Hu-1 monovalent recipients¹⁵. More recently, studies have demonstrated that targeting immune responses to ancestral SARS-CoV-2 by means of additional booster doses increases susceptibility to reinfection with emerging Omicron variants¹⁶. Though these reinfection study designs may be confounded by selection bias¹⁷, reports assessing XBB.1.5 vaccine-induced pseudovirus neutralizing antibodies indeed observe the back-boosting of responses against historic variants that is characteristic of immune imprinting^5,8.

To further evaluate the impact of immune imprinting on XBB.1.5 vaccine immunogenicity, we identified a group of individuals who did...