Hawrusik R, Kawai A, Agan AA, Koram N, Avula R, Brown JS, Burk J, Cai B, Calingaert B, Chomistek AK, Connolly JG, Daniels K, Dea K, DeFor T, DeVries A, Diessner B, Djibo DA, Ezzy SM, Johannes C, Layton JB, Love SAM, Mayer SE, Ma Q, Moyneur E, Nolan MB, Platt R, Rao S, Reynolds JS, Selvan M, Sharma V, McMahill-Walraven CN, Vetter J, Ziyadeh NJ, Gilsenan A, Fuller CC. Incidence rates of safety outcomes in a post-approval study of the Pfizer-BioNTech monovalent COVID-19 vaccine in the United States, booster dose analysis. Poster presented at the 2025 International Society for Pharmacoepidemiology (ISPE) Annual Meeting; August 24, 2025. Washington, DC.

BACKGROUND: The original Pfizer-BioNTech monovalent COVID-19 vaccine (BNT162b2) was authorized for emergency use in the United States (US) from December 2020 – April 2023, including as a booster dose beginning in 2021. A post-approval study is in progress to evaluate the safety of the BNT162b2 vaccine as a booster dose using data from 5 US health insurers participating in the Sentinel System [EUPAS43468].

OBJECTIVES: To describe incidence rates (IR) of myocarditis/pericarditis (myo/peri) in BNT162b2 booster recipients and matched comparators, as well as pooled IRs for 25 additional safety events regardless of exposure status to assess feasibility of a final analysis.

METHODS: Among those aged ≥6 months who previously received ≥2 monovalent BNT162b2 doses December 2020 – April 2023, we matched BNT162b2 booster recipients (3rd dose >60 days after 2nd dose) to those who had not received a booster as of the same calendar date using a 1:2 variable matching ratio by age, sex, US state, calendar time, time since last BNT152b2 dose, and propensity score. Follow-up time was censored on health plan disenrollment, death, or receipt of another COVID-19 vaccine. Algorithms to define 26 safety events and risk windows were informed by studies from the FDA’s BEST Initiative and CDC VSD. In the matched population, IR per 100,000 person years were estimated overall, and in age-defined subgroups (5-11, 12-17, 18-64, 65+ yrs). For myo/peri, IR were stratified by exposure status.

RESULTS: Before matching, we identified 3,136,153 booster recipients and 6,802,363 eligible comparators; 2.1 million booster recipients were included in the matched population. In the overall population, the IR for myo/peri was 45.4 with similar IR in the exposed (47.9) and comparator groups (43.9). Venous thromboembolism [VTE] (IR: 643.1) and acute respiratory distress syndrome (IR: 637.2) were the most common events overall (irrespective of exposure status); < 5 events were observed for acute disseminated encephalomyelitis, Guillain-Barré syndrome, transverse myelitis, anaphylaxis, Kawasaki disease, thrombotic thrombocytopenic purpura and multisystem inflammatory syndrome. The most common events in children were appendicitis (5-11 yrs: 150.2; 12-17 yrs: 215.4) and convulsions (5-11 yrs: 130.7; 12-17 yrs: 157.1); VTE was the most common event in adults (18-65 yrs: 388.5; 65+ yrs 1,695.9). In children 5 to 11 yrs, < 5 events were observed for myo/peri, narcolepsy, deep vein thrombosis, immune thrombocytopenia, and VTE.

CONCLUSIONS: Interim results suggest sufficient sample size to estimate robust effect estimates for many safety events in the final analysis. Precision of hazards ratio estimates for some outcomes may be limited in subgroups defined by age, including for myo/peri in children 5-11 yrs.