RMD Open
Figure 2 Comparison of long-term outcomes between clinical deep remission (CliDR) and non-CliDR groups over 5 years (60 months). (A) Proportion of patients in Disease Activity Score in 28 joints using C reactive protein remission during 5-year post- remission follow-up visits. (B) Kaplan-Meier analysis of survival probability.
Among the covariates included in the multivariable model, family history of rheumatic disease (HR=1.01, 95% CI 0.47 to 2.15; p=0.986) and RF titre (HR=1.00, 95% CI 1.00 to 1.001; p=0.447) were not significantly associated with the outcome. The presence of comorbid- ities showed a trend toward a protective effect, though it did not reach statistical significance (HR=0.61, 95% CI 0.37 to 1.02; p=0.057). DISCUSSION This study demonstrates the prognostic benefits of CliDR for maintaining long-term remission in RA in a real-world setting. We found that patients who achieved
CliDR—defined as the absence of swollen and tender joints combined with normal inflammatory markers— exhibited significantly higher rates of sustained remis- sion over 5 years. Notably, among those undergoing drug tapering, patients who had attained CliDR prior to tapering showed a superior ability to maintain remission compared with those in non-CliDR remission. The treat-to-target (T2T) strategy, supported by current EULAR, ACR and Asia Pacific League of Associations for Rheumatology (APLAR) recommendations, empha- sises sustained remission as a key therapeutic objective to prevent structural damage and optimise long-term outcomes. 1 Current recommendations by EULAR, ACR
Table 2 Univariable and multivariable cox regression analyses of predictors for the first investigator-reported relapse after remission
Univariate analysis
Multivariable analysis
Variable
HR (95%CI)
P value HR (95%CI)
P value
Group (CliDR vs non-CliDR) Tapering status (yes vs no) Age (per 1-year increase) Family history (yes vs no)
0.53 (0.29 to 0.98) 4.42 (2.78 to 7.02) 1.01 (1.00 to 1.03) 2.18 (1.09 to 4.37) 0.66 (0.35 to 1.22) 1.01 (0.97 to 1.04) 1.00 (1.00 to 1.003) 0.72 (0.47 to 1.12) 1.00 (1.00 to 1.001) 1.13 (0.69 to 1.92) 0.53 (0.33 to 0.87)
0.040*
0.57 (0.20 to 1.62) 8.47 (5.01 to 14.32)
0.292
<0.001*
<0.001
0.107 0.028* 0.185 0.663 0.820 0.145 0.075* 0.634 0.012*
1.01 (0.47 to 2.15)
0.986
Sex (female vs male)
Disease duration (per 1-year increase) Anti-CCP titre (per 1-unit increase)
Extra-articular manifestations (present vs absent)
RF titre (per 1-unit increase)
1.00 (1.00 to 1.001)
0.447
Smoking (yes vs no)
Comorbidities (yes vs no)
0.61 (0.37 to 1.02) 0.26 (0.07 to 0.98)
0.057 0.046
CliDR×tapering status (interaction)
Variable coding: For categorical variables, the reference category is the absence of the characteristic (eg, no, absent, male). For continuous variables, the HR corresponds to a one-unit increase. Time-dependent covariate: Tapering status was modelled as a time-dependent covariate (see Methods). The interaction term (CliDR×tapering status) tests whether the effect of tapering differs between groups. In the multivariable model, the HR for CliDR reflects its effect during the non-tapering period (ie, before tapering or among patients who never tapered). *Multivariable model inclusion: Variables with p<0.1 in the univariable analysis were entered into the multivariable model. Anti-CCP, anti-cyclic citrullinated peptide antibody; CliDR, clinical deep remission; RF, rheumatoid factor.
6
Li H, et al . RMD Open 2026; 12 :e006387. doi:10.1136/rmdopen-2025-006387
Powered by FlippingBook