POS1194 RISK OF INCIDENT POLYMYALGIA RHEUMATICA FOLLOWING EXPOSURE TO RECOMBINANT ZOSTER VACCINE IN ADULTS ≥50 YEARS OF AGE IN THE UNITED STATES

Background: Recombinant zoster vaccine (RZV) was approved as a two-dose vaccine for the prevention of herpes zoster for all adults ≥50 years of age in the United States (US) in October 2017. Pre-licensure randomized clinical trials for RZV identified more cases of polymyalgia rheumatica (PMR) among vaccinated individuals (32/14,654) than among placebo recipients (29/14,660) but these studies were not powered to detect a significant difference in PMR incidence. Objectives: This targeted safety study is a post-market regulatory commitment to the US Food and Drug Administration (FDA) and European Medicines Agency to evaluate the risk of new-onset PMR following RZV in adults ≥50 years of age in the US using real-world data. Methods: This retrospective cohort study was conducted using administrative claims data from four large national US health plans and one regional insurer that contribute to the FDA Sentinel System. We compared the risk of incident PMR between RZV-exposed individuals and a random sample of unvaccinated individuals with no prior RZV at the time of a preventive care visit (e.g., annual wellness, mammography, or colonoscopy visit). Cohort sizes were set a priori to power the study to detect a relative risk ≥2.0. The study population included index dates (RZV or preventive care visit) from January 1, 2018 through May 31, 2019; we excluded individuals with an ICD-10-CM diagnosis code for PMR in the 365 days prior to index date. We followed all RZV exposures (Dose 1 and Dose 2) and comparators until the earliest of an incident PMR event, disenrollment, death, receipt of live attenuated zoster vaccine, or the end of 183 days of follow-up. Comparators were censored if they received RZV during follow-up; they then became eligible for the vaccinated cohort. Incident PMR was defined as a PMR diagnosis code followed by a dispensing of oral glucocorticoids within 6 months and another dispensing within 6 months of the first dispensing. The diagnosis code date was considered the event date. Board certified rheumatologists reviewed medical charts from a random sample of potential PMR events, including vaccinated individuals and comparators, to assess the accuracy of this definition. We calculated propensity scores based on demographics, health conditions, and health care utilization assessed during the 365-day pre-index baseline period and performed inverse probability of treatment weighting with 1% trimming to control for confounding. Using Cox proportional hazards regression we calculated the adjusted hazard ratio (HR) of PMR among the RZV-exposed versus comparators, with separate analyses of Dose 1 and Dose 2. We performed a sensitivity analysis on chart-confirmed or -probable cases from a ~25% random sample of our overall analytic cohort from which events were selected for medical record review. Another sensitivity analysis censored follow-up at influenza vaccination, which is a hypothesized risk factor for PMR onset. A post-hoc analysis employed a 66-day post-index lag period (the average time between symptom onset and diagnosis observed in chart review) to avoid capturing PMR with onset before the index date. While all five health plans contributed data for the primary analysis, only four were able to contribute data for the sensitivity and post-hoc analyses. Results: Our analytic sample included 300,648 RZV-exposed individuals and 1,156,964 unexposed comparators. Before weighting, the two cohorts had similar demographic and health history characteristics, although the vaccinated cohort was slightly older (mean age 68.9 vs 67.3 years). All baseline characteristics were balanced after weighting based on standardized mean differences <0.1; the weighted cohorts had a mean age of 68 years, were 59% female, and their most common comorbidities were diabetes, chronic kidney disease, and ischemic heart disease. Post-index influenza vaccination occurred with greater frequency among the vaccinated cohort (Table 1). We identified 127 incident PMR events during follow-up after Dose 1, 68 events after Dose 2, and 777 events in the comparator group. A total of 220 PMR events had charts available for medical record review and 144 had sufficient information for adjudication, of which 75.0% were adjudicated as confirmed or probable cases, supporting the validity of our claims-based PMR definition. In adjusted analyses, we found a 49% lower risk of PMR after RZV Dose 1 (HR 0.51 [95% confidence interval (CI): 0.41- 0.62]) and a 59% lower risk of PMR after RZV Dose 2 (HR 0.41 [95% CI: 0.31-0.55]) compared to unvaccinated comparators (Table 2). An analysis restricted to confirmed and probable adjudicated cases also found a lower risk of PMR among RZV recipients, although with low PMR counts, the Dose 2 result was non-significant (Dose 1 HR 0.37 [95% CI: 0.19-0.74]; Dose 2 HR 0.76 [95% CI: 0.36-1.59]). Results were robust to censoring at influenza vaccination during follow-up (Dose 1 HR 0.51 [95% CI: 0.41-0.64]; Dose 2 HR 0.38 [95% CI: 0.27-0.53]). The post-hoc analysis with a 66-day lag period estimated similar protective effects for Dose 1 (HR 0.48; 95% CI: 0.37-0.62) and Dose 2 (HR 0.52; 95% CI: 0.37-0.74). Conclusion: This study found no evidence of an increased risk of PMR following RZV exposure. Instead, we observed a statistically significant reduced risk of PMR following RZV. These findings accounted for known potential confounders and were consistent across all additional analyses. REFERENCES: NIL . Acknowledgements: The authors would like to thank: Juliane Reynolds, Laura Shockro, Laura Hou, and Katherine Yih from the Harvard Pilgrim Healthcare Institute; Aziza Jamal-Allial and Brian Bennett from Carelon Research; Steve Ezzy, Liz Taggert, Sarah Vosters, and Judy Wong from Optum; the Safety Surveillance and Collaboration Team at CVS Health, Carla Brannan, Elena Cruse, and Vaibhav Sharma from CVS Health; and Valentine Franck and Susan Gamble from GSK. Disclosure of Interests: Sophie Mayer: None declared, Dongdong Li: None declared, Fang Zhang: None declared, Mengqi Cen: None declared, Sarah Alam: None declared, Alexander Peters: None declared, Elizabeth Messenger-Jones: None declared, Richard Platt: None declared, Sebastian Unizony: None declared, Erick Moyneur: None declared, Qianli Ma: None declared, Mano Selvan: None declared, Rachel Ogilvie: None declared, Najat Ziyadeh: None declared, Kimberly Daniels: None declared, Anna Wentz: None declared, Djeneba Djibo: None declared, Cheryl McMahill-Walraven: None declared, Marina Scolnik GSK, Janssen, AstraZeneca, Novartis, Astrazeneca, GSK, Maria Lorena Brance: None declared, Driss Oraichi GSK Employee, GSK Employee, Harry Seifert GSK Employee, GSK Employee, Huifeng Yun GSK Employee, GSK Employee, O'Mareen Spence GSK Employee, GSK Employee, Sheryl Kluberg: None declared. © The Authors 2025. This abstract is an open access article published in Annals of Rheumatic Diseases under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Neither EULAR nor the publisher make any representation as to the accuracy of the content. The authors are solely responsible for the content in their abstract including accuracy of the facts, statements, results, conclusion, citing resources etc.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.