The OMERACT MRI in Arthritis Working Group — Update on Status and Future Research Priorities

Validation of the OMERACT RA MRI Joint Space Narrowing Score (RAMRIS-JSN)

Improved outcome measures of structural joint damage in RA are still needed. The OMERACT RAMRIS has gained acceptance as outcome measure of joint inflammation and bone erosion in RA^1,2,3,4,5,6. However, cartilage damage is an important part of the disease process in RA, and studies assessing JSN on conventional radiography have shown that JSN is independently associated with functional impairment and decreased work ability¹¹. Consequently, incorporating a JSN assessment into RAMRIS will broaden its assessment of structural joint damage, and may also improve the regulatory view of the usefulness of MRI in RA.

Evaluation of cartilage/JSN was left out in the early phase of developing the RAMRIS (late 1990s) because image quality at that time was insufficient. At OMERACT 10 the assessment of joint cartilage/JSN by MRI was reconsidered, based on the fact that image quality has increased, and a preliminary RAMRIS JSN scoring system was subsequently developed based on cross-sectional studies, including comparison with computed tomography^12,13. Before OMERACT 12, validation was performed in a longitudinal study. The RAMRIS-JSN score showed high overall intrareader and interreader reliability, and moderate sensitivity to change¹⁴ supporting its validity as a tool for assessing JSN in RA clinical trials.

Validation of the OMERACT PsA MRI Scoring System (PsAMRIS)

Sensitive outcome measures in PsA, capturing the diverse inflammatory and damage pathologies, are warranted. At OMERACT 9 and OMERACT 10, a PsA MRI scoring system (PsAMRIS, assessing synovitis, tenosynovitis, osteitis, bone erosions, and proliferations and periarticular inflammation), including validation data for hand PsA, was presented^7,8,15. However, further validation was hindered by lack of appropriate datasets given a paucity of MRI outcome studies in PsA. Recently, data from a randomized placebo-controlled trial of patients with PsA¹⁶ were made available to our MRI group, and a validation exercise of the PsAMRIS, based on this material, was undertaken. PsAMRIS showed good intrareader and interreader reliability for most variables in the hand and foot, and the scores of inflammatory features were responsive to change¹⁷. This suggests that PsAMRIS is a valid tool for MRI assessment of hands and feet in PsA clinical trials.

Validation of the OMERACT Hand OA MRI Scoring System (HOAMRIS)

Structure and symptomatology relationships in OA are still poorly understood, and structure modification has been difficult to achieve, in part because of poor outcome measures. Sensitive outcome measures in hand OA are therefore needed, and MRI provides assessment of appropriate pathologies. At an imaging level, established hand OA demonstrates many pathologies in common with hand PsA, and it has been a logical step to extend the OMERACT MRI group’s activities into developing and validating an MRI scoring system for hand OA, given the group’s extensive knowledge and experience in developing definitions and scoring systems for other peripheral joint diseases. Consequently, the OMERACT HOAMRIS was developed and presented at OMERACT 11⁹. However, further validation was needed, including validation of the score in a longitudinal dataset. At OMERACT 12, the responsiveness and longitudinal reliability of the HOAMRIS were tested. Good cross-sectional interreader reliability was found on status scores, whereas the reliability of change scores was variable¹⁸. Results suggest that MRI is sensitive to change in hand OA, but that further validation of MRI measurements is needed before MRI can be recommended as a primary outcome measure in clinical hand OA trials.

Future Research

At OMERACT 12 potential future research areas were discussed during the MRI special interest group session. In RA, further knowledge of the performance of the RAMRIS-JSN component score in randomized controlled trials and in comparison with conventional radiography is important. Focus on which particular sites show more responsiveness to change, so that the scoring system can be potentially reduced in terms of number of anatomical sites evaluated, is also relevant because this could improve its feasibility.

Several approaches to further developing the RAMRIS may also be considered, including validation in the first finger, the interphalangeal joints, and the feet. Further, tenosynovitis is a frequent finding in RA, and the participants at OMERACT strongly suggested investigation of the potential value of MRI assessment of tenosynovitis as an addition to the RAMRIS, with testing of reliability, responsiveness, and discriminatory ability of any putative tenosynovitis score. Peripheral enthesitis, which is a common pathology in spondyloarthritides, was also suggested as an area of future research.

Regarding HOAMRIS, further testing in longitudinal exercises using images from higher field strengths was suggested. With respect to PsAMRIS, gathering data from other randomized controlled trials in PsA is crucial to fully clarify the utility of PsAMRIS. Participants at OMERACT 12 also felt that the development and provision of a PsAMRIS atlas, either electronically or Web-based, would further improve the utility of the tool. It was also suggested that development of educational tools, such as online training modules to allow knowledge transfer regarding both RAMRIS and PsAMRIS, would be helpful. Finally, investigations of alternative MRI methodologies such as whole-body MRI, dynamic contrast-enhanced MRI, and diffusion-weighted MRI, were also suggested as potential areas for future focus for the group.

Important new data on MRI in RA, PsA, and hand OA validating the RAMRIS-JSN, the PsAMRIS, and the HOAMRIS methods, were presented at OMERACT 12, as was a series of exciting potential future research priorities. The group will continue its efforts to optimize the value of MRI as a useful tool in rheumatology clinical trials.