Retention rates of different Janus kinase inhibitors in rheumatoid arthritis: experience from a large monocentric cohort.

The efficacy of Janus kinase inhibitors (JAKi) in rheumatoid arthritis (RA) has been clearly shown. However, information on comparative drug retention rates (DRRs) of different JAKi is heterogeneous. The aim of this study was to compute and compare DRRs of different JAKi in a large cohort of RA patients. Patients with RA treated with at least one JAKi and followed up at our centre were retrospectively identified. DRRs of each JAKi were computed at 24 months. The association of baseline features with drug persistence was tested. Variations in 28-joint Disease Activity Score-C-reactive protein (DAS28-CRP) and Clinical Disease Activity Index (CDAI) scores between baseline and 12 months were analysed. The study included 365 patients, with a total of 463 therapy courses. Tofacitinib was the most prescribed JAKi (33%), followed by baricitinib (25%), upadacitinib (24%), and filgotinib (21%). The mean treatment duration was 24 ± 17 months, with a maximum of 70 months. At 24 months, the overall DRR was 86%. DRRs were not significantly different across different JAKi. The only baseline predictor of treatment discontinuation was previous treatment with a biological disease-modifying anti-rheumatic drug (bDMARD) (hazard ratio 1.65, 95% confidence interval 1.08-2.53; p = 0.021). There were significant reductions in DAS28-CRP and CDAI 1 year after treatment start. In our large, monocentric cohort, the overall 24 month DRR for JAKi was greater than 80%. No significant differences in retention were found among different JAKi. Persistence was lower in patients who had previously been treated with other bDMARDs.