Background and Importance The therapeutic armamentarium for rheumatoid arthritis (RA) has been remodelled over the last decades with the advent of biologic disease-modifying antirheumatic drugs (bDMARDs) and the emergence of Janus Kinase inhibitors (JAKi). So far, real-world data comparing the persistence of these different treatment approaches are scarce.
Aim and Objectives This study aims to compare treatment persistence between JAKi and bDMARDs in a real-world setting of RA patients.
Material and Methods A retrospective study (January 2017 to September 2022), including all RA patients from a tertiary hospital under treatment with JAKi, tumour necrosis factor inhibitor (TNFi), interleukin (IL) 6 inhibitor (IL6i), cluster of differentiation (CD) 80/86 inhibitor (CD80/86i), or CD20 inhibitor (CD20i). Demographic, clinical, and pharmacological data were collected from hospital claim records. Persistence was examined through Kaplan-Meier survival analysis. Median survival times were compared statistically using log-rank test and Cox model. Statistical analyses and graphic representations were performed utilising STATA15® software.
Results We included 582 cases: 166 (28.5%) JAKi treatments, 180 (30.9%) TNFi treatments, 124 (21.3%) IL6i treatments, 64 (11.0%) CD80/86i treatments, and 48 (8.3%) CD20i treatments, corresponding to 293 RA patients (86% women, 63 ± 14 years old).
The median JAKi treatment persistence was 428 [95 CI% =262–609] days, which did not differ significantly with regard to the median treatment persistence of: TNFi (HR=1.19 [95 CI%=0.91–1.56]; p=0.215), IL6i (HR=1.06 [95 CI%=0.79–1.43]; p=0.695), CD80/86i (HR=1.40 [95CI%=0.99–1.98]; p=0.054), and CD20i (HR=0.77 [95 CI%=0.50–1.18]; p=0.227). Median treatment persistences are presented in table 1.
Kaplan-Meier curves represent the estimated survival functions (figure 1).
Conclusion and Relevance Based on the results from our RA real-world cohort, JAKi treatment persistence is in line with TNFi and other bDMARDs treatment persistences. Further research is needed to confirm our findings.
Conflict of Interest No conflict of interest