Association between prior HBV vaccine responsiveness and booster-induced anti-SARS-CoV-2 spike antibody titers following the third BNT162b2 dose in Japanese healthcare workers: A prospective observational study

Inter-individual variability in vaccine-induced humoral responses has been reported even among generally healthy individuals. In our previous study, healthcare workers with poor hepatitis B virus (HBV) vaccine responsiveness were more likely to show a low anti-SARS-CoV-2 spike antibody response after the first BNT162b2 dose. This study examined whether such differences persisted after a third BNT162b2 dose. This was an extended follow-up study of a previously reported prospective observational cohort of Japanese healthcare workers. The primary outcome was the quantitative anti-SARS-CoV-2 spike antibody titer measured 12 weeks after the third BNT162b2 dose. Participants were classified as normal HBV vaccine responders, weak responders, or non-responders. Comparisons across HBV response groups were performed using the Kruskal-Wallis test and multivariable linear regression adjusted for age and sex. A low antibody-response analysis (<15 U/mL) after the first dose was evaluated as a secondary analysis. Among 741 participants with antibody measurements 12 weeks after the third dose, 486 were normal HBV vaccine responders, 229 were weak responders, and 26 were non-responders. The median anti-SARS-CoV-2 spike antibody titer was 891,900 U/mL. Titers did not differ significantly across HBV response groups in the unadjusted analysis (P = .3718) or after adjustment for age and sex (global P = .3971). Compared with normal responders, the adjusted geometric mean ratio was 0.94 (95% CI, 0.83-1.07) for weak responders and 0.83 (95% CI, 0.60-1.15) for non-responders. By contrast, after the first dose, HBV vaccine non-responders had an increased risk of a low antibody response (<15 U/mL) compared with normal responders (adjusted RR, 1.85; 95% CI, 1.10-3.13). Differences in early anti-SARS-CoV-2 spike antibody responses associated with prior HBV vaccine responsiveness were no longer statistically apparent 12 weeks after the third BNT162b2 dose. These findings suggest convergence of measured humoral antibody responses after booster vaccination, although they do not establish that the underlying biological mechanisms of vaccine hyporesponsiveness were eliminated.