The past two decades have seen unprecedented advances in therapy for rheumatoid arthritis (RA), with nine effective biological agents introduced since 1998. Each of these therapies gained regulatory approval based on radiographic evidence of suppression of bone erosion and articular cartilage loss (measured as joint-space narrowing) in randomized, controlled clinical trials (RCT). Recent trends, however, have made radiography increasingly impractical for monitoring disease progression and treatment response in clinical trials. As a result, pharmaceutical developers have turned to MRI as a potential alternative, based on MRI’s ability to discriminate treatment efficacy in less time and with fewer patients than radiography can. To date, MRI has been successfully used in more than ten RCTs involving nine different therapeutic agents. All of these studies used the same scoring method, the OMERACT RA MRI Score (RAMRIS). The original draft of RAMRIS, however, included only synovitis, osteitis and bone erosion. It did not include assessment of cartilage loss or joint-space narrowing (JSN). This limitation has been an obstacle to substituting MRI for radiography in clinical trials, as articular cartilage loss is at least as important, if not more, than bone erosion is in determining long-term disability in RA. Moreover, suppressing bone erosion does not always ensure that cartilage loss has been suppressed as well.
MRI is excellent for evaluating cartilage, and unlike radiography, which can only estimate cartilage thickness indirectly based on joint-space width, MRI can visualize the articular cartilage directly. This obviates problems associated with projectional superimposition of the rims of joints over joint spaces, which can mimic JSN on radiographs, and erroneous preservation of joint space despite cartilage loss, due to ligamentous insufficiency or interposition of synovial tissue or joint fluid.
A number of scoring methods have been developed to evaluate cartilage loss in the hands and wrists in clinical trials of RA. These include a 9-point scale developed by Peterfy, et al. in 1998, a 5-point scale developed by McQueen, et al in 2010, and a 5-point scale developed by the OMERACT MRI group as an extension to RAMRIS in 2011. Similar to radiographic scoring methods, the 5-point scales focus on JSN; however, the 9-point MRI cartilage loss scale evaluates articular cartilage thickness directly. All of these scores have shown strong correlation with radiographic JSN. The 9-point MRI cartilage loss scale has furthermore been used successfully in several RCTs. In the SCORE trial of rituximab plus methotrexate versus placebo plus methotrexate, the 9-point scale showed significant suppression of cartilage loss by rituximab at 24 weeks (the earliest time point examined) and 52 weeks. These changes on MRI at 24 and 52 weeks correlated with radiographic JSN at 52 weeks. In the JADA trial of baricitinib, the 9-point MRI scale demonstrated significant suppression of cartilage loss within 12 weeks of treatment with 4mg or 8mg of baricitinib plus methotrexate compared to treament with lower doses of baricitinib plus methotrexate or with placebo plus methotrexate. Finally, in a multicenter RCT of infliximab plus methotrexate versus placebo plus methotrexate, the 9-point scale showed significant suppression of cartilage loss by infliximab by 14 weeks (the earliest time point examined). The MRI technique used in each of these studies was the same as that used for monitoring bone erosion with RAMRIS: fat-suppressed, T1-weighted 3D gradient-echo, which has been shown to delineate articular cartilage accurately in various joints, including the MCPs, and is commercially available on all clinical MRI systems operating at magnetic field strengths of 1.0 T or higher. Moreover, the acrylic frame used in these studies to reproducibly fix the position of the bones in the hand, wrist and fingers across serial MRI examinations also fixes the degree of flexion of each joint facilitating accurate assessments of change. Thus, MRI protocols do not need to be modified or extended in order to add MRI Cartilage Score to assessments of joint damage.
In conclusion, MRI can demonstrate suppression of not only bone erosion but cartilage loss in RCTs of RA with far fewer patients and much less time than needed to do so with radiography. With the recent shift towards active-comparator study designs, which require longer study durations and more patients to demonstrate therapeutic superiority, and the increasing difficulty in recruiting RA patients into clinical trials, there is a growing need for more sensitive methods, such as MRI, to offset the escalating costs, patient exposure and logistical challenges associated with these trends.
Disclosure of Interest C. Peterfy Shareholder of: Spire Sciences, Consultant for: AbbVie, Amgen, Articulinx, AstraZeneca, Bayer, BioClinica, Bristol Myers-Squibb, Celgene, Centocor, Genentech/Roche, Icon Medical Imaging, Jannsen, Lilly, Medimmune, Merck/Schering-Plough, Moximed, Novartis, Perceptive Informatics, Pfizer, UCB, VirtualScopics, Employee of: Spire Sciences