Responsiveness of different dynamic contrast-enhanced magnetic resonance imaging approaches: a post-hoc analysis of a randomized controlled trial of certolizumab pegol in rheumatoid arthritis

Objective: The aim was to explore dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) as an early marker of therapeutic response in patients with rheumatoid arthritis (RA) starting treatment with certolizumab pegol (CZP). Method: In 40 RA patients initiating CZP (27 patients) or 2 weeks of placebo (PCB) followed by CZP (13 patients), DCE-MRI of the metacarpophalangeal (MCP) and proximal interphalangeal (PIP) joints was performed at weeks 0, 1, 2, 4, 8, and 16. Using semi-automated software, three methods for drawing volume regions of interest (ROIs) in MCP2-5 and PIP2-5 were applied: 'Standard' (slices: all; joints: MCP2-5 together and PIP2-5 together), 'Detailed' (slices: slices with high-quality visualization; joints: as Standard), and 'Single-joint' (slices: as Detailed; joints: each joint separately). The number of enhancing voxels (Nvoxel), initial rate of enhancement (IRE), and maximum enhancement (ME) were extracted and analysed for each method. Results: Nvoxel in MCP2-5, and IRE and ME in PIP2-5 decreased statistically significantly (Wilcoxon rank-sum test, p < 0.02-0.03) after 16 weeks of treatment for the Standard method. Nvoxel and ME decreased significantly more in the CZP group than in the PCB group after 1 week of treatment, but not at later time-points. There were no significant changes for DCE-MRI parameters for the Detailed and Single-joint methods. Conclusions: Certain DCE-MRI parameters detected decreased inflammation during CZP treatment in RA patients. Using specific criteria for ROIs, as in the Detailed and Single-joint methods, decreased the statistical power and could not show any changes over time.