Abstract
As COVID-19 transitions into endemicity and vaccines are annually updated to circulating SARS-CoV-2 lineages such as JN.1, exposure intervals and immune imprinting become critical considerations for vaccination strategy. Imprinting by the ancestral spike protein has been observed with the bivalent Wuhan-Hu-1/BA.4-5 vaccine and its persistence can be further evaluated in the context of the more recent XBB.1.5 monovalent vaccine. We assessed antibody responses in individuals who received three to four doses of Wuhan-Hu-1, one dose of bivalent Wuhan-Hu-1/BA.4-5, and one dose of XBB.1.5 vaccine (bivalent recipients). We compared these to individuals who received three to four doses of Wuhan-Hu-1 and one dose of XBB.1.5 vaccine without prior bivalent vaccination (bivalent non-recipients). Before XBB.1.5 vaccination, bivalent non-recipients demonstrated decreased breadth and potency of neutralizing antibodies compared to recipients, but at post-vaccination exhibited greater boosting of neutralizing antibodies against XBB.1.5 (18.4X versus 6.2X), EG.5.1 (30.9X versus 7.0X), and JN.1 (9.2X versus 3.7X) variants with trends toward higher neutralizing titers and comparable variant cross-neutralization. Increased boosting in non-recipients were similarly observed for IgA and total IgG/A/M isotypes binding the spike receptor-binding domain but not IgG nor IgM. Bivalent non-recipients had longer intervals between exposures, which has been reported to enhance antibody boosting; however, bivalent receipt and interval were tightly linked variables, preventing the isolation of individual contributions to boosting. Nonetheless, significant "back-boosting" of ancestral SARS-CoV-2 titers upon XBB.1.5 vaccination in both participant groups indicate that immune imprinting continues to affect contemporary vaccines. Altogether, our findings highlight imprinting and exposure intervals as important phenomena underlying variant-adapted COVID-19 vaccine immunogenicity.