LAUSR

Background Dynamic contrast-enhanced MRI (DCE-MRI) has been proposed for evaluating treatment response in RA. In a 16-week anti-TNF trial, DCE-MRI measures of inflammation analyzed for regions of interest (ROIs) covering MCP joints 2–5 and PIP joints 2–5 detected improvements at week 16, but not at earlier time point1. Objectives To investigate if solely analyzing joints fulfilling predefined MRI quality criteria for joint visualization would increase the responsiveness and discrimination between treatments of DCE-MRI. Methods Patients with active RA despite stable DMARD therapy for ≥12 weeks were randomized 2:1 to certolizumab pegol (CZP) or 2 weeks of placebo (PBO) followed by CZP (CZP+PBO). MRIs were obtained at weeks 0 (baseline), 1, 2, 4, 8 and 16. Only joints fulfilling MRI joint quality criteria (≥3MCP/≥2PIP joint slices including the distal and/or the proximal bone of the joint and part of the joint cavity) were included in analyses. ROIs covering each joint were analyzed for number of enhancing voxels (Nvoxel), initial rate of enhancement (IRE) and maximum enhancement (ME) using the DYNAMIKA software (Image Analysis, UK). Results For 38 (CZP: 26; PBO+CZP: 12) of the 40 randomized patients, ≥1 joint fulfilled the quality criteria at baseline. 31 MCP2, 28 MCP3, 23 MCP4, 7 MCP5, 29 PIP2, 29 PIP3, 28 PIP4 and 12 PIP5 joints were included. No individual joints showed significant changes over time or differences between groups. Analyses by joint group (MCP2–4 and PIP2–4) had few data available. Nvoxel and ME decreased numerically, but not significantly, for PIP2–4.Table 1. Baseline values of and changes in DCE-MRI parameters for MCP2–4 and PIP2–4 Baseline Change week 0–week 1 Change week 0–week 2 Change week 0–week 4 Change week 0–week 8 Change week 0–week 16 Median change [number] PBO+CZP/CZP PBO+CZP/CZP PBO+CZP/CZP PBO+CZP/CZP PBO+CZP/CZP PBO+CZP/CZP Nvoxel  MCP2–4 295/131 NA/-503 62/0 NA/0 90/53 NA/-473 [5]/[12] [1]/[4] [1]/[5] [2]/[4] [1]/[5]  PIP2–4 108/144 -35/0 -56/-12 -80/-75 -69*/10 -32/-12 [8]/[15] [3]/[6] [5]/[7] [5]/[4] [7]/[10] [5]/[8] IRE  MCP2–4 0.000/0.004 NA/0.000 0.006/0.00 NA/0.000 0.273/-0.023 NA/0.002 [4]/[9] [0]/[3] [2]/[2] [0]/[4] [2]/[2] [0]/[5]  PIP2–4 0.013/0.006 -0.002/-0.001 0.000/0.001 0.003/0.000 0.014*/0.000 0.000/0.023 [8]/[10] [2]/[3] [5]/[5] [5]/[3] [5]/[5] [4]/[2] ME  PIP2–4 0.013/0.006 -0.002/-0.001 0.000/0.001 0.003/0.000 0.014*/0.000 0.000/0.023 [8]/[10] [2]/[3] [5]/[5] [5]/[3] [5]/[5] [4]/[2]  PIP2–4 1.76/1.99 -0.16/-0.30 -0.17/0.02 -0.23/0.03 -0.02/-0.01 -0.19/-0.12 [8]/[10] [2]/[3] [5]/[5] [5]/[3] [5]/[5] [4]/[2] NA: Not applicable. Difference between time points: Wilcoxon Signed Ranks test: *p<0.05. Conclusions There were no statistically significant changes in DCE-MRI on joint level or joint group level or between groups. Applying strict joint coverage quality criteria compromises the statistical power of the DCE-MRI analyses underlining the importance of standardization of the method. References Østergaard. Arthritis Rheum 2014;66(Suppl. 10):518. Disclosure of Interest M. B. Axelsen Grant/research support from: UCB Nordic funded the study, H. Bliddal Grant/research support from: UCB Nordic, L. T. H. Jacobssen Consultant for: Abbvie, Celegen, MSD, Novartis and UCB, M. Hansen: None declared, A. Dudek: None declared, M. Rell-Bakalarska: None declared, M. Boesen Consultant for: Chairman of the clinical advisory board Image Analysis, LTD London, J. Stefanek: None declared, B. Sundman-Engberg: None declared, M. Østergaard Grant/research support from: and/or speaking/consultant fees: Abbvie, BMS, Boehringer-Ingelheim, Celgene, Eli-Lilly, Centocor, GSK, Hospira, Janssen, Merck, Mundipharma, Novartis, Novo, Orion, Pfizer, Regeneron, Schering-Plough, Roche, Takeda, UCB, and Wyeth