Presented at the 41st Annual International Society for Pharmacoepidemiology Conference
Objective: To rapidly monitor the incidence proportions of 10 prespecified safety outcomes following ABRYSVO™ to comparator vaccination (influenza, Tdap, COVID-19 vaccines) in pregnant individuals using US administrative data.
Design: ABRYSVO™ is a bivalent respiratory syncytial virus (RSV) prefusion F vaccine approved in the US for pregnant individuals at 32 0/7 through 36 6/7 weeks’ gestation to prevent RSV-associated lower respiratory tract illness in infants. A non-interventional safety study (including both a rapid cycle analysis [RCA] and a cohort study; EUPAS1000000115) was initiated to rapidly monitor ABRYSVO safety. Cumulative data through the second of five planned hypothesis tests in the RCA results are presented.
Method: Pregnancies vaccinated at 32-36 weeks gestation with ABRYSVO (exposure) or influenza/Tdap/COVID-19 (comparator) were identified from September 2018-January 2023 (historical comparator) and September 2023-February 2024 (exposure, concurrent comparator) at 5 national and regional insurers. Outcome incidence was assessed using maximized Sequential Probability Ratio Tests, and included preterm birth, hypertensive disorders of pregnancy [HDP] (with and without secular trend adjustment), stillbirth, premature labor without preterm delivery, premature rupture of membranes [PROM], preterm PROM, Guillain-Barré syndrome [GBS], small for gestational age, large for gestational age, and low birth weight. Potential confounding variables included in covariate stratification were maternal age, season, gestational age at vaccination, and evidence of high-risk for preterm birth.
Results: Historical Comparison: 5,027 ABRYSVO-exposed and 143,459 comparator pregnancies were observed. Among evaluated outcomes, only PROM (crude incidence proportion [IP] 13.4% [95% CI: 12.4-14.3%] vs. 12.5% [12.4-12.7%]) and HDP (without secular trend adjustments) (crude IP 17.4% [16.4-18.5%] vs. 13.4% [13.2-13.6%])) met the criteria to reject the null hypothesis of no excess risk (a statistical signal).
Concurrent Comparison: 5,028 ABRYSVO-exposed and 6,528 comparator pregnancies were observed. Among evaluated outcomes, only HDP (crude IP 17.4% [16.4-18.5%] vs. 15.7% [14.8-16.6%]) met the criteria to reject the null hypothesis of no excess risk (a statistical signal).
Monitoring is ongoing for all remaining outcomes that have not reached statistical stopping points.
Conclusions: Further investigation is needed to explore a potential increased risk of PROM and HDP following ABRYSVO exposure. RCAs, while allowing for early detection of safety signals, are limited in their ability to adequately control for confounding. These data, while informative to identify potential safety risks, are therefore not sufficient to demonstrate causality.
