Prior herpes zoster occurrence and high-dose corticosteroids increase herpes zoster risk in rheumatoid arthritis patients receiving janus kinase inhibitors in a retrospective and observational study.

ABSTRACT

Herpes zoster (HZ) risk is increased in rheumatoid arthritis (RA) patients receiving Janus kinase inhibitors (JAKi) therapy. Identifying and evaluating the risk factors of HZ development in patients receiving JAKi therapy would be clinically helpful. We investigated HZ's incidence rates (IR), identified the risk factors, and further assessed their influence on HZ development in RA patients undergoing JAKi therapy. We retrospectively evaluated 249 RA patients who received JAKi therapy between 2015 and 2023. Data regarding clinical characteristics, HZ reactivation, HZ vaccination status, and concomitant medication use were collected. Among 249 JAKi-treated patients, 44 developed new-onset HZ (tofacitinib, 28/142; baricitinib, 6/35; upadacitinib,10/72), with an IR of 5.11/100patient-years. Multivariate analysis revealed significant predictors of HZ development a long JAKi exposure period, prior HZ or COVID-19 history, and concomitant high-dose corticosteroids use. The interval between JAKi initiation and HZ development was significantly shorter in patients with prior HZ history than in those without (median, 6.5 months versus 33.5 months, p < 0.001), suggesting "biphasic" emergence of HZ. Only one patient who had experienced an HZ episode while receiving JAKi developed recurrent HZ. None of the seventeen patients immunized with the non-live recombinant zoster vaccine developed HZ. Our JAKi-treated patients had elevated HZ risks, a class effect across different JAKi. A long exposure period, prior history of HZ or COVID-19, and concomitant high-dose corticosteroid treatment may further increase the risk. The emergence of HZ shows a biphasic pattern early HZ development in patients with prior HZ and late development in those without. Key Points • An increased risk of HZ was observed in Taiwanese RA patients treated with JAKi, presenting as a class effect. • Patients with a long JAKi exposure period, prior history of HZ or COVID-19, and concomitant use of high-dose corticosteroids were at high risk of HZ while receiving JAKi therapy. • The interval between JAKi initiation and HZ occurrence was shorter in patients with prior HZ than in those without, showing "biphasic" emergence.