(SLC-052) Safety Evaluation of Multiple Vaccination Regimens for the Elderly in Japan: The VENUS Study

Background: Verifying vaccine safety requires addressing confounding, selection, and information biases. Target Trial Emulation (TTE) framework has recently been proposed and utilized for interpreting causal effects for treatment efficacy in real-world data setting. The framework may be useful when assessing the safety of multiple vaccine regimens. Several vaccines, including COVID-19, influenza, and pneumococcal vaccines, are recommended in older adults, however sometimes, the simultaneous vaccination would be a concern. Thus, evaluating the safety of different vaccine regimens is increasingly important.

Objectives: This study aims to implement and evaluate analytical methods based on the TTE framework to assess the safety of multiple vaccine regimens, focusing on influenza and pneumococcal vaccinations in the elderly.

Methods: A population-based cohort study was conducted using medical claims data and a vaccine registry provided by a city. Participants were individuals aged 65 years or older who had received Pneumococcal vaccine. We compared adverse events (e.g. cellulitis) among the following four vaccine regimens. Adverse events were defined using ICD-10 codes and drug prescriptions for outcome definitions.
Regimen 1: Pneumococcal vaccine only.
Regimen 2: Simultaneous vaccination of pneumococcal vaccine and influenza vaccine
Regimen 3: Pneumococcal vaccine followed by influenza vaccine within one month
Regimen 4: Pneumococcal vaccine followed by influenza vaccine within one to two months later
These regimes were compared using a three-step TTE method (cloning / censoring / weighting) from the vaccination date, estimating adjusted rate ratios (aRR) with 95% confidence intervals by Poisson regression model. Also, a cohort approach treating vaccines as time-varying exposures was applied.

Results: Among the individuals in the city, 52,078 individuals received the pneumococcal vaccine, and 523,267 received the influenza vaccine between 2014 to 2023. For those receiving the influenza vaccine one month after the pneumococcal vaccine, an immortal time existed where adverse events could not occur. The TTE-based three-step method effectively addressed this bias. In contrast, the time-varying exposure approach did not adequately account for immortal time in multi-vaccine regimen safety assessments.

Conclusions: The three-step TTE method (cloning / censoring / weighting) is a useful approach for evaluating the safety of multiple vaccine regimens.