Background: Recently, the disease activity of rheumatoid arthritis (RA) was improved due to the ‘treat-to-target’ strategy. However, some patients remain various symptoms despite recommended treatment was performed. Then, the term of ‘difficult-to-treat RA (D2TRA)’ is widely recognized. It is unknown how the difference of type of biological disease-modifying anti rheumatic dugs (bDMARDs)/Janus kinase inhibitor (JAKi) will affect clinical efficacy in patients with D2TRA. Moreover, the risk factor to inhibit the clinical response in patients with D2TRA is unknown.

Objectives: The aim of this study was to evaluate the treatment response in patients with D2TRA who were treated with interleukin 6 receptor inhibitor (IL-6Ri), abatacept and JAKi.

Methods: This study used the multicenter database included 673 RA patients treated with bDMARDs/JAKi (tocilizumab 240, sarilumab 67, abatacept 146, tofacitinib 101, baricitinib 83, upadacitinib 20, peficitinib 14, filgotinib 2). Two hundred forty-two patients were treated as first line bDMARDs/JAKi (IL-6Ri 117, abatacept 63, JAKi 62), 211 patients were treated as second line bDMARDs/JAKi (IL-6Ri 117, abatacept 37, JAKi 57), 220 patients were treated as third and more bDMARDs/JAKi. In these 220 patients, 82 patients did not meet D2TRA criteria (IL-6Ri 42, abatacept 15, JAKi 25) and 138 patients met D2TRA criteria (IL-6Ri 31, abatacept 31, JAKi 76). In all patients, we analyzed 138 patients with D2TRA (113 female, mean age was 63.1 ± 13.7 years). Drug retention rate and effectiveness of bDMARDs/JAKi in patients with D2TRA were evaluated for 24 weeks. Multivariate linear regression analysis was performed to clarify the risk factors to inhibit the clinical response.

Results: Drug retention rate of patients with D2TRA at 24 weeks was 67.7% in IL-6Ri group, 74.2% in abatacept group, 61.8% in JAKi group. Drug retention rate in patients with D2TRA was not different between groups (IL-6Ri vs abatacept: p=0.86, IL-6Ri vs JAKi group: p=0.39, abatacept vs JAKi group: p=0.33). DAS28-CRP at 4, 12, 24 weeks decreased in all group ( Figure 1 ). Abatacept showed lower improvement ratio of DAS28-CRP at 24 weeks compared to IL-6Ri group (IL-6Ri vs abatacept: p<0.01, IL-6Ri vs JAKi: p=0.1, abatacept vs JAKi: p=0.07). Good responder (defined as decrease in DAS28-CRP score > 1.2 with a score < 3.2) was 52.4% patients in IL-6Ri, 17.4% patients in abatacept, 29.8% patients in JAKi. SDAI and CDAI at 4, 12, 24 weeks decreased in all group ( Figure 1 ). There were no diferences between the groups in improvement ratio of SDAI (IL-6Ri vs abatacept: p=0.11, IL-6Ri vs JAKi: p=0.81, abatacept vs JAKi: p=0.08) and CDAI (IL-6Ri vs abatacept: p=0.31, IL-6Ri vs JAKi: p=0.82, abatacept vs JAKi: p=0.13) at 24 weeks. HAQ was 1.42, 1.15, 1.39 at baseline, 1.27, 1.07, 1.22 at 4 weeks, 1.17, 1.07, 1.17 at 12 weeks, 1.26, 1.06, 1.14 at 24 weeks in IL-6Ri group, abatacept and JAKi, respectively. Multivariate linear regression analysis revealed that high HAQ (β=0.28, p=0.02) and high dosage of glucocorticoid (β=0.67, p<0.01) inhibited the improvement of DAS28-CRP. Type of bDMARDs/JAKi (β=-0.09, p=0.36) did not affect the DAS28-CRP improvement for 24 weeks.

Conclusion: Drug retention rate and clinical efficacy of D2TRA patients were not different among IL-6Ri, abatacept and JAKi. DT2RA patient with functional disorder and high dosage of glucocorticoid were risk factor to inhibit the clinical response.

Disclosure of Interests: None declared