Background Achieving remission is essential for avoiding joint destruction and disability in patients with rheumatoid arthritis (RA). Tumor necrosis factor inhibitors (TNFi) therapies represent an important advancement in therapy for RA. However, there remains a proportion of patients who do not improve despite TNFi therapies. Recently, Terao et al. (1) analyzed the distribution of affected joints in the 28 joints in patients with RA and showed that RA patients can be categorized into three subgroups based on their affected joints (1: PIP joints dominant, 2: MCP joints dominant, and 3: large joints with wrist joints dominant). Nevertheless, it remains obscure whether these three subgroups were correlated with treatment response.

Objectives To analyze the prognostic significance of data collected at starting TNFi, especially the distribution of affected joints, to predict remission in RA patients at week 16.

Methods Data from 62 TNFi-treated patients with RA at the baseline were used as variables to predict remission. The disease status at the baseline and week 16 was assessed using the disease activity score (DAS 28) and patients achieving remission at week 16 were identified according to EULAR criteria (DAS 28 <2.6). The mean age of patients was 57.6 and the mean disease duration was 9.7 years. The mean disease activity at the baseline was 4.93 and 22.7 for DAS28 and simplified disease activity index (SDAI), respectively. The mean modified health assessment questionnaire (mHAQ) was 0.658. Thirty-eight patients (61%) were naïve to biologic agents. Fifteen (24%), 16 (26%), and 31 (50%) patients were treated with etanercept, adalimumab, and golimumab, respectively. Forty-eight (77%) and 30 (48%) patients were treated with methotrexate (MTX) and glucocorticoids (GC) concurrently. All of the patients were classified into 3 groups according to their affected joints which have tenderness or swelling, that is, 1: PIP joints dominant (PIPd) group, 2: MCP joints dominant (MCPd) group, and 3: large joints with wrist joints dominant (L-Wd) group. The correlation of baseline characteristics with achievement of remission at week 16 was explored by multivariate logistic regression analysis.

Results The patients were classified into 11 (18%) of PIPd group, 19 (31%) of MCPd group, and 32 (52%) of L-Wd group according to their affected joints. Univariate analyses revealed that patients who achieved remission showed significantly lower levels of DAS28, mHAQ, ESR, and CRP, and shorter disease duration than those without remission. It was also disclosed that patients in PIPd group were more likely to achieve remission than patients in MCPd group or L-Wd group. The baseline characteristics including concurrent use of MTX and/or GC, previous biologics therapy, tender joint count, and swollen joint count were not associated with response to TNFi. Among the baseline characteristics, lower levels of ESR and belonging to PIPd group were independently correlated with achievement of remission in multivariate analysis (p=0.012 and 0.028, respectively).

Conclusions We suggested that RA patients predominantly affected with PIP joints were likely to achieve remission in response to TNFi. The distribution of affected joints could predict future response to TNFi in RA patients.

References

1. Terao C et al. PLoS One. 2013; 8: e59341.

Disclosure of Interest None declared

DOI 10.1136/annrheumdis-2014-eular.1980