1266: Time-course Ultrasonography to Predict Flare After Discontinuation of Biologics in Patients with Rheumatoid Arthritis Achieving Stringent Remission

Takehisa Ogura¹, Takaharu Katagiri², Yuto Takakura³, Chihiro Imaizumi⁴ and Hideto Kameda⁴, ¹Toho university, Meguro-ku, Japan, ²Toho University School of Medicine (Ohashi Medical Center), Division of Rheumatology, Department of Internal Medicine, Tokyo, Japan, ³Toho University, Setagayaku, Japan, ⁴Toho University, Tokyo, Japan

Background/Purpose: Ultrasonography (US) has been suggested to be useful in predicting flare in patients with rheumatoid arthritis (RA) after discontinuation of biological disease-modifying antirheumatic drugs (bDMARDs). This study aimed to investigate whether US can predict flare after discontinuation of bDMARDs in RA patients who have achieved stringent remission criteria.

Methods: We prospectively enrolled RA patients who maintained a simplified disease activity index ≤ 3.3 and discontinued bDMARDs and measured clinical assessment and US every 2-3 months for 2 years. The US examination was performed on 40 joints using the semi-quantitative method of 0-3 on the Grey-scale (GS) and Power Doppler (PD), and the total values for each patient were used as the GS score and PD score. Joints graded as GS score ≥ 2 or PD score ≥ 1 were counted as US arthritis. In addition, tendons at 36 sites were counted with or without tendinitis/tenosynovitis to obtain a tendon score.

Results: Thirty-six patients were enrolled and two patients who dropped out early without flare were excluded from the comparative analyses. At baseline, the median GS score was 7, PD score was 0, US arthritis was 0, and tendon score was 0. The total PD score was 0 in 26 patients (72%) and it was 1 in 5 patients (14%). There were no significant differences in US findings between the relapse group (20 patients) and the non-relapse group (14 patients). Although a Cox proportional hazards model with baseline US findings showed no significant findings, using a time-dependent Cox proportional hazards model to analyze time-course US findings, the PD score and the tendon score were significant with hazard ratio of 1.14 (95% CI, 1.08-1.21) and 1.37 (95% CI, 1.13-1.66), respectively. However, among the relapse group, 11 in 20 patients (55%) had a PD score of 0 on US finding before relapse.

Conclusion: The US findings suggest that an elevated PD and tendon score are associated with flare after bDMARDs discontinuation, although it is insufficient for the flare prediction.

Disclosures: T. Ogura, AbbVie/Abbott, Bristol-Myers Squibb(BMS), Eli Lilly, Janssen, Eisai, Asahi kasei; T. Katagiri, None; Y. Takakura, None; C. Imaizumi, None; H. Kameda, AbbVie, Asahi-Kasei, Lilly, Janssen, Novartis, Sanofi, Bristol-Myers Squibb, Chugai, Mitsubishi Tanabe, Eisai, Astellas, Gilead, Taisho.