Presented at the 41st Annual International Society for Pharmacoepidemiology Conference
Background: Pivotal clinical trials for recombinant zoster vaccine (RZV), a two-dose vaccine to prevent herpes zoster,identified more cases of supraventricular tachycardia (SVT) among recipients of RZV (n=6) versus placebo(n=0). The present study fulfills a regulatory commitment to evaluate the risk of SVT in real-world data.
Objectives: To evaluate the relative risk (RR) of incident SVT following RZV using a self-controlled risk interval (SCRI)design.
Methods: This study used US administrative claims data from five health plans in the FDA’s Sentinel distributed datanetwork. We included adults aged ≥50 years receiving RZV between Jan 1, 2018 – Dec 31, 2019 withenrollment in medical and drug coverage from 365 days before through 60 days after RZV.
The SCRI analysis included individuals with SVT in the immediate 30-day risk window (RW) following RZV or asubsequent 30-day control window (CW). Incident SVT was defined as a diagnosis during an inpatient oremergency department visit with no SVT diagnoses in the prior 365 days. Claims-based cases wereadjudicated by a cardiologist using medical records.
A temporal scan evaluated clusters of SVT during follow-up versus the null hypothesis that cases follow auniform distribution. Our primary analysis evaluated the RR of chart-confirmed SVT in the RW versus CWfollowing any dose, estimated using conditional Poisson regression. We used the RR to calculate theattributable risk (AR) per 100,000 doses. Sensitivity analyses included probable or asymptomatic SVT (SVT < 6minutes detected on routine monitoring) along with confirmed cases. A post-hoc analysis excluded caseswith a medical record history of SVT >1 year prior to their event.
Results: Among 522,461 adults receiving RZV, we identified 350 incident SVT events within 60 days using claims.Temporal scan identified a cluster of 14 cases in days 1-5 after RZV relative to 5.25 expected cases over 5days (p=0.04). A total of 176 (50.3%) SVT cases had available medical records, adjudicated as: 64 confirmed, 7probable, 6 possible, 17 asymptomatic, 25 insufficient, and 57 ruled out. Eight confirmed cases (7 RW, 1 CW)had a distant (>1 year) history of SVT. The RR of confirmed SVT in the RW (n=40, 62.5%) versus CW (n=24,37.5%) was 1.66 (95% CI: 0.98, 2.89), corresponding to an AR of 7.06 excess cases per 100,000 doses (95% CI:-0.36, 11.61). The effect was smaller when including probable (RR 1.53; 95% CI: 0.93, 2.56) or asymptomaticcases (RR 1.25; 95% CI: 0.79, 1.99) and in post-hoc analysis excluding cases with history of SVT (RR 1.43; 95%CI: 0.82, 2.56).
Conclusions: This study found evidence suggestive of an increased risk of SVT following RZV exposure, although the RRwas not statistically significant. Main limitations include a low chart retrieval rate and lack of statistical power.
