POS0675 IMPACT OF PAST USE OF DISEASE MODIFYING ANTI-RHEUMATIC DRUGS ON JAK INHIBITOR TREATMENT FOR RHEUMATOID ARTHRITIS - DATA FROM THE FUKUI ISHIKAWA TOYAMA DATABASE OF RHEUMATOID ARTHRITIS

Abstract

BackgroundCurrently, five types of Janus kinase inhibitors (JAKis) are used for rheumatoid arthritis (RA) treatment. The number of cases in which multiple JAKis have been prescribed is increasing. However, the real-world efficacy and safety of JAKis and related factors require further evaluation.ObjectivesThe primary objective of this study was to elucidate the impact of past use of disease-modifying anti-rheumatic drugs on RA treatment using JAKis. The secondary objective was to investigate the safety profiles of these agents in a real-world setting.MethodsOf the 303 JAKi-treated patients in the Fukui Ishikawa Toyama Database of RA included in this study, 30 had switched from one JAKi to another (JJ group), 214 switched from a biologic agent to a JAKi (BJ group), and 47 were naïve to either biologics or JAKis (NJ group). We compared baseline factors, treatment response, and JAKi continuation rates among the three groups. Factors related to JAKi discontinuation were assessed using Cox regression analysis. Furthermore, we investigated adverse events and reported them using exposure-adjusted incidence rates (EAIR; incidence rates per 100 patient-years).ResultsData from the 303 cases were analyzed (mean age = 63.6 years; female, 82%; mean RA duration, 176 months). Of the 303 patients, 118, 106, 50, and 29 were treated with tofacitinib, baricitinib, peficitinib, and upadacitinib, respectively, on initial observation. Rate of concomitant use with methotrexate and prednisolone was 52% and 49%, respectively.Regarding efficacy, no significant differences were observed among the three groups in terms of treatment response and JAKi continuation rates, except for the 6-month treatment response between the JJ and NJ groups. Cox regression analysis of the 303 cases revealed that only past use of JAKis during the disease history was significantly associated with JAKi discontinuation. The Kaplan–Meier method showed that patients who previously used JAKis had significantly shorter median JAKi treatment duration than those without such a history (20.9 vs. 54.7 months; p = 0.012). Treatment response was significantly poor in patients who had previously used JAKis, especially 6 months after treatment initiation.In terms of safety, the total exposure period for the 303 cases was 495 person-years, and the total number of adverse events was 161 (EAIR, 32.5). There were 12 cases (EAIR, 2.5) of serious infections, 23 cases (EAIR, 5.1) of herpes zoster, 7 cases (EAIR, 1.4) of malignant tumors, and 4 cases (EAIR, 0.8) of MACE. Adverse events led to JAKi discontinuation in 34 patients (EAIR, 6.9); the main causes of adverse events leading to treatment discontinuation were infectious diseases in 10 cases (EAIR, 2) and neoplasms in 4 cases (EAIR, 0.8). Within 1 year of initiating JAKi therapy, 21 patients discontinued treatment owing to adverse events, which accounted for 27% of the reasons for treatment discontinuation.We also investigated cases of JAKi dose reduction, observed in 42 of the 303 cases. Among them, 10 patients required a re-increase in the JAKi dose, and 13 patients (56.5%) were able to maintain the reduced dose for more than 1 year. The remaining 19 patients were excluded from the analysis because the treatment duration at the lower dose had not exceeded 1 year at the time of data extraction. No difference in disease activity at the time of dose reduction was observed between those who maintained the new dose and those who did not (mean DAS28-CRP: 1.48 ± 0.26 vs. 1.89 ± 0.62).ConclusionPast use of JAKis may contribute to decreased response and continuation rates for JAKi treatment. In this study, conducted in Japan, development of herpes zoster was found to the most frequent adverse event among the priority survey items.AcknowledgementsI have no acknowledgements to declare.Disclosure of InterestsNone declared