(066) Multi-regional population-based cohort study for the evaluation of the association between herpes zoster and mRNA vaccinations for SARS-CoV-2: The VENUS Study

Publication Number: 403

Background: Despite many observational studies having been conducted worldwide, the increased risk of herpes zoster (HZ) due to COVID-19 vaccines remains controversial.

Objectives: To assess the increased risk of first-time HZ with the use of mRNA vaccines against COVID-19.

Methods: We performed a population-based cohort study in four municipalities in Japan. We included residents, aged over 11 years on October 1, 2020 (Cohort Entry Date: CED). We excluded residents who met at least one following criteria: (1) those who had SARS-CoV-2 infection prior to CED, (2) those who had a history of HZ prior to CED, and (3) those who have been enrolled for insurance for less than 1 year prior to CED. The follow-up period was from CED until the earliest date of one of the following: (1) HZ occurrence date, (2) November 30, 2021, (3) date of death, (4) the day before a sample that indicated a positive SARS-CoV-2 test result, and (5) the day before having the ChAdOx1 vaccination. The vaccination status of each individual was categorized according to vaccine type (BNT162b2 or mRNA-1273) and dose (unvaccinated/within 28 days of the first dose or from the first vacation date until the second vaccination date/after day 29 of the first vaccination until the second vaccination/within 28 days of the second dose/after day 29 of the second dose). The onset of HZ was defined as a first prescription of anti-virus drugs with an indication for HZ in the same month as the diagnosis of HZ (ICD10 code: B02). The HZ incidence rate after vaccination was compared with the unvaccinated period. We considered background information, such as age, sex, city, frequency of clinic/hospital visits, comorbidities, and medications as covariates. The adjusted incidence rate ratio (IRR) and 95% confidence interval (CI) per each period were estimated using a conditional Poisson regression model including the vaccination status as a time-dependent covariate.

Results: We identified 339,548 individuals for the cohort, which yielded 377.438 person-years. The median age was 74 years (interquartile range: 66-81 years), and 57.8% were female. During the follow-up, 296,242 individuals (87.2%) completed primary immunization, most were homologous vaccination of the BNT162b2 vaccine. The adjusted IRR (95%CI) within 28 days of the first and second BNT162b2 vaccinations was 1.04 (0.82, 1.30) and 1.06 (0.87, 1.30), respectively. An increased risk of HZ was noted after the second BNT162b2 vaccination in the women under 50 years old subgroup. There were no cases of HZ following the mRNA-1273 vaccination.

Conclusions: The results showed that there was no increased HZ risk after the BNT162b2 vaccination in the overall population. However, the effect modification of age and sex was observed. The assessment on mRNA-1273 was unable due to the small sample size.