Background/Purpose: Tofacitinib is an oral Janus kinase inhibitor for the treatment of RA. Tofacitinib was approved in December 2014 for RA under Taiwan’s National Health Insurance (NHI) reimbursement system. A drug-based registry (XTRA; established July 2016) collects real-world short- and long-term effectiveness, safety, treatment patterns, and persistence in RA patients newly prescribed tofacitinib or a TNF inhibitor (TNFi).

Methods: This observational study within the XTRA Registry comprises a 24-month enrollment plus 36-month follow-up period (August 2016 to August 2021). Patients receive standard care for RA according to the treating rheumatologists and NHI reimbursement criteria (Disease Activity Score 28-ESR [DAS28-ESR]>5.1; failure of 2 DMARDs including MTX; no tuberculosis). The tolerance and effectiveness of tofacitinib were assessed using Work Productivity and Activity Index-RA (WPAI-RA), HAQ-DI, DAS28-ESR, and Clinical Disease Activity Index (CDAI) scores at baseline and 24-week intervals. Comparisons with TNFi used Chi-square test for categorical variables and Student’s t test for continuous variables, with no adjustment for multiplicity.

Results: As of January 2018, 211 patients were enrolled in the registry (n=113 tofacitinib; n=98 TNFi [etanercept, adalimumab, or golimumab]). Demographic characteristics were similar between tofacitinib initiators and TNFi initiators. TNFi initiators had higher baseline median DAS28-ESR and CDAI scores (6.16 and 36.6, respectively) vs tofacitinib initiators (5.74 and 28.7, respectively) (p<0.05).  Tofacitinib initiators comprised 69.9% who had not previously received TNFi. More tofacitinib initiators (30.1%) than TNFi initiators (2.0%) received ≥1 previous TNFi (p<0.05). Persistence at the end of the first year was similar for tofacitinib initiators (88.5%) and TNFi initiators (90.8%). Numerically decreased functional impairment and improved productivity over time measured by WPAI-RA were seen for both tofacitinib and TNFi initiators. At Week 24, a similar proportion of tofacitinib initiators (23.3%) achieved CDAI low disease activity (LDA; 2.8<CDAI≤10) vs TNFi initiators (20.8%); similar effectiveness was also shown for tofacitinib in terms of DAS28-ESR<3.2 (LDA; 20.7% vs 18.0%). At Week 48, TNFi initiators and tofacitinib initiators achieved CDAI LDA (37.0% and 31.8% of patients), DAS28-ESR<2.6 (remission; 7.4% and 4.2%), and DAS28-ESR LDA (29.6% and 16.7%), respectively. The incidence rate of all-cause adverse events (AEs) was numerically higher in tofacitinib initiators vs TNFi initiators (44.9 vs 33.1 events/100 patient‑years).

Conclusion: This drug-based registry is an effective tool for collecting real-world data from RA patients. Preliminary data showed meaningful effectiveness and patient-reported outcomes in both tofacitinib and TNFi initiators. The incidence and types of AEs in this real world setting under the same reimbursement criteria in the 6 medical centers across Taiwan were comparable with the known safety profile of tofacitinib. Additional data from this registry will provide pertinent information on these advanced therapies among patients with severe RA.