Background: Recombinant zoster vaccine (RZV), a 2-dose series vaccine, has been shown to be ≥ 90% effective against herpes zoster (HZ) in clinical trials. However, evidence from real-world data is limited.
Objectives: To evaluate real-world RZV effectiveness against HZ overall, by age groups, by corticosteroid use status, by time between doses 1 and 2 and by time-since-vaccination.
Methods: We conducted a prospective cohort study from January 1, 2018, through June 30, 2022, in persons aged ≥ 50 years at 4 integrated healthcare systems in the Vaccine Safety Datalink (VSD) in the US. The outcome was incident HZ, defined as an HZ diagnosis (ICD10 code B02) together with an antiviral prescription. We used Cox regression to compare the hazard of HZ in fully vaccinated (≥30 days after dose 2) and partially vaccinated (≥30 days dose after 1) persons with unvaccinated persons, adjusted for demographics, comorbidities, prior zoster vaccine live status, VSD site and calendar time. Vaccine effectiveness (VE) was calculated as one minus the adjusted hazard ratio. VE was also estimated by age at first dose, by corticosteroid use, by time between doses 1 and 2, and by time since last dose.
Results: The study included 1.9 million persons who contributed 6.9 million person-years (PY) of follow up. Of the 42,554 incident HZ cases, 5% received ≥ 1 RZV dose. The crude incidence of HZ per 1000 PY was 6.7 among unvaccinated, 2.6 among partially vaccinated, and 1.8 among fully vaccinated persons. After adjustment, VE of 1 dose was 64% (95% confidence interval [CI] 61% – 66%) and VE of 2 doses was 76% (95% CI 75% - 78%). Following 1 dose only, VE was 70% during the first year, 44% during the second year, and 50% after the second year. Following the second dose, VE was 78% during the first year, 75% during the second year, and 74% after the second year. Among fully vaccinated persons, VE was 80% among persons who received their first dose at ages 50-64 years and 75% among those who received their first dose at ages ≥65 years; VE was 66% in persons who received corticosteroids before their first or second dose compared to 77% in persons who had not received corticosteroids before vaccination; VE was similar for persons who received their second dose ≤ 6 months after their first dose (76%) compared with persons who received their second dose > 6 months to < 1 year (77%) or ≥1 year (76%) after their first dose.
Conclusions: In a real-world setting, we found that two doses of RZV were effective, but VE estimates were lower than those reported by clinical trials. One dose of RZV was initially effective but one-dose effectiveness decreased substantially after the first year. Two-dose effectiveness did not substantially decrease during the study period.
