HAND-DOMINANT JOINT INVOLVEMENT PATTERN ASSOCIATES WITH FAVORABLE, AND POLYARTHRITIS WITH UNFAVORABLE, TREATMENT RESPONSE TO BOTH csDMARDs AND bDAMRDs IN EARLY RHEUMATOID ARTHRITIS: A COMBINED ANALYSIS OF NORD-STAR AND BeSt TRIALS

Background:
We recently reported the existence of four joint involvement pattern (JIP) subgroups of treatment-naive early rheumatoid arthritis (RA) [1]. Among these, hand-dominant JIP (JIP-Hand) patients, who exhibit a hand-predominant arthritis distribution, demonstrated higher success rates with methotrexate and higher remission rates. It is unknown whether these phenotypic clusters are associated with differential treatment effects in other cohorts and treatments as well.
Objectives:
To investigate whether these JIP-subgroups are predictive of disease activity following conventional synthetic DMARDs (csDMARDs) and biological DMARDs (bDMARDs) treatment, and to compare the effect of JIP-subgroups with other clinical parameters in treatment-naive early RA.
Methods:
An individual patient data pooled analysis was conducted using two randomized controlled trials, the NORD-STAR [2, 3] and BeSt trials [4]. JIP-subgroup assignment was based on previously identified subgroups using baseline clinical characteristics, primarily joint involvement in the 66/68 joint scheme. The primary outcome was the longitudinal CDAI measured up to week 48. Secondary outcomes included CDAI remission, DAS28-CRP, and HAQ. The associations between JIP-subgroups and other clinical predictors were evaluated using linear mixed model analysis, adjusting for baseline CDAI difference as covariates and using random intercepts for the patients. We also examined whether the impact of JIP-subgroups varied with combined bDMARD treatments through an interaction term analysis.
Results:
We included a total of 1,250 treatment-naïve early RA patients (Table 1, Figure 1A). Patients with a hand-dominant JIP (JIP-Hand) showed significantly better CDAI scores after treatment, even with correction for the baseline CDAI difference (Beta on CDAI = -1.4 [95% CI -2.3 – -0.55], p = 0.0016), indicating that JIP-Hand patients showed a 1.4-point better improvement in CDAI after treatment compared to the other three JIP-subgroups, averaged over all time points (Figure 1B-D). In contrast, patients with a polyarthritis pattern (JIP-Poly) exhibited worse outcomes (Beta = 0.95 [0.064 – 1.8], p = 0.035). Female sex was associated with worse CDAI values (Beta = 1.2 [0.40 – 2.0], p = 0.0031), whereas anti-citrullinated protein antibodies did not show a significant association (Beta = 0.19 [-0.69 – 1.1], p = 0.67) (Figure 1E-G). Analysis of secondary outcomes also confirmed the robust association between JIP-Hand and improved disease activity outcomes after treatment, with JIP-Hand being associated with higher likelihood of achieving CDAI remission (OR = 1.7 [95% CI 1.2 – 2.4], p = 0.0054). When compared as groups, csDMARDs and combined bDMARDs were similarly effective for the respective JIP-subgroups (interaction p > 0.10).
Conclusion:
In early RA, csDMARD and bDMARD treatments resulted in the greatest improvement in disease activity in the hand-dominant JIP and the least improvement in the polyarthritis JIP.