Background Patients with rheumatoid arthritis (RA) are increasingly achieving clinical remission with intensive treatment regimens. Ultrasound subclinical synovitis are reported in remission states. The presence of power doppler (PD) signal in these states is a predictor of radiographic progression and flare. The multi-biomarker disease activity (MBDA) blood test reflects the overall disease activity in RA and its relationship with US needs further study.
Objectives To evaluate associations among US, serum biomarker concentrations and MBDA score in the REMIRA cohort.
Methods We studied 95 patients with RA from the REMIRA cohort who at study entry, had been on stable therapy for ≥6 months with DAS28-ESR ≤3.2 for at least 1 month. DAS28-ESR was assessed and serum concentrations of VCAM-1, EGF, VEGF-A, IL-6, TNF-RI, YKL-40, MMP-1, MMP-3, leptin, resistin, SAA, and CRP were measured every 3 months from baseline (BL) to 1 year. MBDA scores were calculated using the validated Vectra® DA algorithm (range 1–100). For each patient, 10 MCP joints and both wrists were studied by US at BL and 1 year, with each joint scored 0–3 for PD and grey scale (GS) to compute total PD and GS scores (range 0-36). More stringent alternative scores (PD‡, GS‡) were computed by using higher thresholds for US signal, based on the distributions of baseline PD/GS signals obtained for MCP joints of patients who had no swollen joints (SJC28=0) for 1 year. Correlations between US scores and DAS28, MBDA score and their components were evaluated by Spearman's rank correlation.
Results Mean (±SD) values for patient characteristics at BL were: age 57 (±14) years, disease duration 50 months (±31), DAS28-ESR 2.1 (±1.0), MBDA score 30 (±13), PD score 3.7 (±5.1), GS score 12.1 (±4.7), PD‡ 1.1 (±2.7), and GS‡ 0.4 (±1.1). 61% were female and 87% were RF+ and/or anti-CCP+. At BL, 67% of patients were in DAS28 remission, 43% were in MBDA remission (defined as ≤25), 31% had PD‡ >0 (91% had PD>0) and 21% had GS‡>0 (100% had GS>0). Among the 62 patients in DAS28 remission, 44% had MBDA score>25, 28% had PD‡>0 (92% had PD>0), 19% had GS‡>0 (100% had GS>0). Significant correlations were observed at BL and 1 year between PD and MBDA scores, DAS28-ESR, and SJC28 (Table). Significant correlations with PD were also observed at BL and 1 year for 4 MBDA biomarker components: CRP (r=0.21 & 0.31), SAA (r=0.21 & 0.35), IL-6 (r=0.26 & 0.34) and MMP-3 (r=0.24 & 0.36). Similar correlations were found with PD‡ at BL and 1 year: CRP (r=0.26 & 0.30), SAA (r=0.27 & 0.33), IL-6 (r=0.39 & 0.31) and MMP-3 (r=0.28 & 0.33). The likelihood of having MBDA score ≤25 was significantly greater in patients with PD‡=0 than in patients with PD‡>0 (52% vs. 24% at BL and 36% vs. 9% at 1 year). There was weak or no significant correlation between GS or GS‡ and DAS28-ESR, MBDA or their components at BL and 1 year.
Conclusions In a cohort of RA patients with low disease activity, MBDA scores and some of its component biomarkers correlated significantly with PD signal but not significantly GS. These results suggest that the MBDA score and its biomarkers can detect low-grade inflammation and subclinical synovitis in patients with RA in low clinical disease activity.
Disclosure of Interest M. Ma Grant/research support: NIHR, T. Garrood: None declared, W. Li Shareholder of: Crescendo Bioscience, Employee of: Crescendo Bioscience, N. Defranoux Shareholder of: Crescendo Bioscience, Employee of: Crescendo Bioscience, G. Kingsley: None declared, D. Scott: None declared, A. Cope: None declared