Immune imprinting as a barrier to effective COVID-19 vaccines

Abstract

Wang and colleagues show that immune imprinting impairs neutralizing antibody titers for bivalent mRNA vaccination against SARS-CoV-2 Omicron subvariants. Imprinting from three doses of monovalent vaccine can be alleviated by BA.5 or BQ-lineage breakthrough infection but not by a bivalent booster.¹

Wang and colleagues show that immune imprinting impairs neutralizing antibody titers for bivalent mRNA vaccination against SARS-CoV-2 Omicron subvariants. Imprinting from three doses of monovalent vaccine can be alleviated by BA.5 or BQ-lineage breakthrough infection but not by a bivalent booster.¹

Main text

The concept of “immune imprinting,” also known as “original antigenic sin,” is not new; it originated in the 1960s with the investigation of immune responses to infection by divergent strains of influenza viruses. Scientists found that people had stronger neutralizing antibody responses toward influenza strains that they had been exposed to during childhood, while more antigenically distinct strains stimulated less effective immune responses.² Mechanistically, when different strains of a pathogen infect the host cell, the immune system dedicates most of the response toward recalling the immune effectors used for the original exposure as opposed to generating responses against the new strains, thus resulting in immune evasion.

Immune imprinting can be induced either separately or concomitantly by one of two mechanisms: one is that the immune system favors a recalled response over a de novo one (“antigenic seniority”), and the other is that the de novo response is actively suppressed (“primary addiction”). Evidence for both phenomena has been found for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection and vaccination.³ Antigenic seniority has been demonstrated by the fact that upon infection of vaccinated individuals with variants such as Alpha, Delta, or Omicron, the neutralizing antibody response remains biased toward wild-type spike. Additionally, infection with SARS-CoV-2 in unvaccinated individuals has been shown to favorably induce antibodies that cross-react with common cold betacoronaviruses HKU1 and OC43. Direct evidence for primary addiction has been demonstrated by experiments that tracked the induction of memory versus de novo B cell responses upon boosting with a BA.1 spike, which resulted in a dramatic suppression of a de novo response in favor of a response biased toward wild-type spike.³ These findings emphasize the strong influence of immune imprinting on the ability to mount effective responses against antigenically variable SARS-CoV-2 variants.

Currently, one of the biggest challenges faced during the COVID-19 pandemic is the maintenance of vaccine efficacy as SARS-CoV-2 continues to evolve. The end of 2021 was marked by the emergence of the Omicron BA.1 variant, which was characterized by dramatic escape of neutralizing antibodies stimulated by two-dose monovalent mRNA vaccines.⁴ Since then, the virus has evolved into numerous Omicron subvariants, including the most recent XBB variants. Each of these variants has exhibited mounting immune escape,⁵ which led to the decision to reformulate mRNA vaccination measures. In September 2022, the bivalent formulations of the Pfizer and Moderna mRNA vaccines were made available. These vaccines included the BA.4/5 spike in addition to the ancestral spike that was present in the monovalent formulations to help boost immune responses toward Omicron subvariants.

While the inclusion of the BA.4/5 spike in the bivalent formulations has helped boost the immune response toward Omicron subvariants, neutralizing antibody titers were nowhere near what was exhibited for wild-type/D614G spike, with all Omicron subvariants, including BA.4/5, still exhibiting dramatic reductions in neutralizing antibody titers.⁶ These were the first indicators of immune imprinting stimulated by the initial doses of monovalent vaccine. Since BA.4/5, there has been the emergence of more immune-evasive lineages including BQ.1, BQ.1.1, BA.2.75, and CH.1.1, as well as the recombinant subvariants XBB.1.5, XBB.1.16, XBB.2.3, EG.5, and FLip. In this issue of Cell Reports Medicine, Wang et al. sought to corroborate the idea of immune imprinting by demonstrating the role of breakthrough infection of Omicron BA.5 or BQ variants in helping overcome immune imprinting.¹ Breakthrough infection represents another means of boosting the immune response toward the pathogen in question, and it has been established that breakthrough infection with SARS-CoV-2 can stimulate a neutralizing antibody response biased toward the infecting variant.⁷

The data presented by Wang et al. demonstrate that immune imprinting is a critical issue in the current course of vaccines. They show that individuals who had three doses of monovalent vaccine with a bivalent booster containing BA.4/5 had little increase in titer toward Omicron subvariants, especially XBB.1.5 and XBB.1.16, relative to individuals who received three doses of monovalent vaccine with a monovalent booster lacking BA.4/5. In contrast, individuals who had received at least two doses of monovalent vaccine and experienced breakthrough infection with either BQ or, to a lesser extent, BA.5 exhibited more noticeable increases in titer toward Omicron XBB subvariants. Using antigenic mapping analysis, they discovered that breakthrough infection minimized antigenic distances between D614G, BA.5, and BQ.1.¹ The antigenic mapping method was originally developed to quantitatively measure the antigenic differences between influenza strains based on hemagglutination neutralization assay results.⁸ More recent results from other studies have also confirmed the concern of immune imprinting in COVID-19 vaccine by demonstrating impaired neutralizing antibody responses against the most recent XBB variants EG.5.1 and FLip.^9,10 Together, these works are crucial in informing vaccination strategies against SARS-CoV-2 and have led to the decision to roll out a monovalent XBB.1.5 mRNA vaccine in fall 2023. Based on the studies by Wang et al. and others, it has become clear that the exclusion of wild-type spike from the Omicron-lineage spike vaccine should help overcome the immune imprinting caused by the earlier vaccine course and better protect against the current Omicron subvariants in circulation.

Despite these advances, surveillance and characterization of new SARS-CoV-2 variants remains critical. The chances are that XBBs will not be the last variants of SARS-CoV-2 that we will face, making studies such as that presented by Wang et al. timely and important in the maintained control of the COVID-19 pandemic.