Reduced versus standard dosing of Janus kinase inhibitors in rheumatoid arthritis: a propensity score-matched real-world study accounting for poor prognostic factors.

Abstract

This study aimed to compare 24-month drug retention between standard-dose and reduced-dose Janus kinase inhibitors (JAKi) in rheumatoid arthritis (RA) and to examine whether poor prognostic factor (PPF) burden and renal function modify treatment persistence, with primary inference restricted to patients with preserved renal function, in whom reduced dosing reflects discretionary clinical decision-making. We conducted a multicentre observational study using a Japanese real-world registry of patients with RA who initiated JAKi between 2013 and 2025. Patients were classified according to their initial JAKi dosing strategy. Primary analyses were performed in patients with preserved renal function. Propensity score matching was used to adjust for confounding by indication. Drug retention over 24 months was assessed using Kaplan-Meier analyses and Cox proportional hazards models, with prespecified stratification by PPF burden (0-2 vs 3-4). Supportive analyses were conducted in patients with reduced renal function. Among 1135 patients, 24-month drug retention was comparable between standard-dose and reduced-dose JAKi in the matched cohort with preserved renal function. Disease activity improved similarly in both groups, and glucocorticoid use remained low without compensatory escalation. Reduced dosing performed comparably in patients with low PPF burden, whereas standard-dose initiation showed more favourable persistence in patients with multiple PPFs. In patients with reduced renal function, standard dosing was associated with higher drug retention. In patients with preserved renal function, reduced-dose JAKi initiation was associated with 24-month treatment persistence comparable with standard dosing. Findings in PPF-stratified analyses and in patients with reduced renal function should be interpreted cautiously and warrant further investigation.