Background Dynamic contrast-enhanced (DCE) magnetic resonance imaging (MRI) has been validated in rheumatoid arthritis for measuring inflammation, but has rarely been studied in psoriatic arthritis (PsA).

Objectives To evaluate changes in DCE-MRI parameters of inflammation in patients with PsA in an investigator-initiated trial, during adalimumab therapy.

Methods Anti-TNF-naïve patients with PsA according to Moll and Wrights criteria, ≥3 swollen joints, ≥3 tender joints, hand involvement (swelling and/or dactylitis) and clinical indication for anti-TNF therapy were included. Conventional and DCE-MRI (0.6 T) was performed at baseline, and after 6 and 48 weeks. All patients received adalimumab 40 mg eow. The PsA MRI scoring system (PsAMRIS) was used for analyzing conventional MRIs, and Dynamika software (Image Analysis Ltd., Leeds, UK) for DCE-MRIs. Regions of interest (ROIs) were drawn around the 2.-5. MCP joints, excluding large blood vessels. The ROIs were used for computing the number of pixels with plateau and washout pattern (N_p+w), the initial rate of enhancement (IRE), and the maximum enhancement. PsAMRIS total was calculated by adding the components of synovitis, flexor tenosynovitis, periarticular inflammation and bone marrow oedema.

Results Patient characteristics (n=9) were: 5 males (56%), median (range) age 39 (25-63) years, joint/skin disease duration 7 (2-59)/12 (2-59) years, 76-swollen joint count (SJC) 20 (6-32), 78-tender joint count (TJC) 41 (9-71), HAQ-score 1.3 (1.1-2.1) and dactylitis count 6 (0-13). Five patients discontinued the trial between 3 and 7 month after inclusion due to lack of efficacy (1 patient), adverse events (2 patients), or violation of study protocol (2 patients). Significant reductions in N_p+wwere observed between baseline and follow-up in the entire patient group (n=9, see table).

DCE-MRI scans were also available for 12 patients who did not have D-MRI performed at inclusion (total 21 patients). For these 21 patients N_p+w, IRE and ME*N_p+w correlated with PsAMRIS synovitis (Spearman’s rho 0.52, -0.51, 0.53, all P<0.05), and with PsAMRIS total (Spearman’s rho 0.60, -0.57, 0.61, all P<0.05). We found no statistical significant correlation between dynamic parameters and DAS28, SJC or TJC.

Conclusions In active PsA patients treated with adalimumab, dynamic MRI parameters decrease significantly as assessed by the Dynamika software, and are associated with findings on conventional MRI. DCE-MRI might be a valuable method for measuring joint inflammation in PsA.

Disclosure of Interest None Declared