MRI-guided treat-to-target (T2T) strategies are no more effective than conventional T2T at improving remission rates or reducing radiographic progression among patients with rheumatoid arthritis (RA) in remission, data from a randomized controlled trial show.
On the basis of the study’s findings, MRI-guided therapy is not recommended for these patients, write Signe Møller-Bisgaard, MD, from the Copenhagen Center for Arthritis Research, Rigshospitalet, in Glostrup, Denmark, and colleagues. They report their findings in an article published online February 5 in JAMA.
Researchers designed the IMAGINE-RA randomized trial, which compared MRI-guided T2T with clinical T2T, in the hope of finding a way to prevent the progressive joint damage that occurs in 20% to 30% of patients with RA who are in clinical remission.
Møller-Bisgaard told Medscape Medical News, “We could not demonstrate any benefit of using an MRI treat-to-target strategy, targeting absence of bone marrow edema for the primary outcomes of achieving DAS28-CRP remission [disease activity score in 28 joints–C-reactive protein <32 with no swollen joints] and eliminating radiographic progression at 24-month follow-up compared to the conventional treat-to-target approach in rheumatoid arthritis patients in clinical remission.”
Bone marrow edema is a strong independent predictor of joint damage progression. Thus, the researchers were surprised that despite achieving a significantly lower bone marrow edema score in the MRI-T2T group, they did not see a slowing of erosive joint progression. Møller-Bisgaard speculated that the follow-up time might have been too short or that bone marrow edema in RA patients in remission might be associated with an overall poorer prognosis that might not be modifiable by treatment.
Arthur F. Kavanaugh, MD, professor of medicine and director of the Center for Innovative Therapy at the University of California, San Diego, told Medscape Medical News that he found the data “pretty convincing.” Kavanaugh, who was not involved in the study, said, “It does not appear that pushing a T2T strategy using MRI results in clinical benefit. I was honestly a little surprised by the results.”
IMAGINE-RA was a 2-year, randomized, multicenter trial conducted at nine hospitals in Denmark. The study enrolled 200 patients with RA in clinical remission (defined as DAS28-CRP <3.2 and no swollen joints). Patients were randomly assigned to undergo MRI-guided T2T (n = 100) or conventional T2T (n = 100) and were assessed every 4 months.
Despite random assignment, at baseline, the MRI group had a lower rate of DAS28-CRP remission compared with the conventional T2T group (86% vs 96%), shorter median disease duration (9.0 vs 11.0 years), higher disease activity on patient assessment (using a visual analogue scale [VAS] of 0–100, 15.0 vs 9.0), more pain (VAS, 15.0 vs 9.0), and more fatigue (VAS, 19.5 vs 12.0). In addition, more patients in the MRI group experienced bone marrow edema (47% vs 36%).
The treatment goal in the MRI group was clinical remission plus an absence of MRI bone marrow edema. The treatment goal in the conventional T2T group was clinical remission. Both groups were treated according to a standard algorithm for escalation of conventional and biologic treatment.
The authors report that at 24 months, there was no difference between the MRI-T2T and conventional T2T groups for reaching the primary clinical outcome of DAS28-CRP remission (85% vs 83%) or the primary radiographic endpoint (66% vs 62%).
Møller-Bisgaard said that the between-group imbalance in baseline characteristics, especially the difference in presence of remission at baseline, might have affected the rates of achieving the primary clinical endpoint but is unlikely to have markedly changed the results. Kavanaugh agreed that the effect would have likely been small.
The protocol included intra-articular glucocorticoid injections to every clinically swollen joint both during and between planned visits, which is standard clinical practice in Denmark. Møller-Bisgaard said this might have contributed to the very low disease activity in both groups, making it harder to demonstrate a difference between groups.
More Treatment in MRI-T2T Group
Serious adverse events (SAEs) occurred in 17% of the MRI-T2T group and in 6% of the conventional T2T group. Treatment intensification was dictated by the study algorithm and was intensified 173 times in 73% of the MRI-T2T group vs 22 times in 17% of the conventional T2T group. At 24 months, 46% of the MRI-T2T patients received biologic treatment vs 2% of the conventional T2T patients.
“We saw significantly more treatment escalations in the MRI-treat-to-target group compared to the conventional-treat-to target group, and significantly more patients received biologic treatment at 24 months,” Møller-Bisgaard said. “The higher rate of SAEs in the MRI-treat-to-target group was likely related to the more intensive therapy administered in this group, and as consequence, more patients in the MRI-treat-to-target group received more medicine with no effect on the primary outcomes but effect on some secondary outcomes.” This included significant improvements in responses to the health assessment questionnaire and in MRI target variable osteitis scores.
Kavanaugh commented that this might in part suggest possible overtreatment in the MRI group. “There was certainly a fairly marked difference in the treatments used by the end of the study,” he said.
In a linked editorial, Daniel Aletaha, MD, from the Division of Rheumatology, Department of Medicine, Medical University of Vienna, Austria, and Josef S. Smolen, MD, from the Department of Internal Medicine, University of Vienna, and the Department of Rheumatology, Vienna General Hospital, Austria, comment, “Altogether, the use of imaging to guide therapy in RA did not meet the criteria for ‘rational use of medicines’ as defined by the World Health Organization because using imaging to guide therapy led to prescription of potentially harmful medicines without differences in the primary outcomes, but at high costs and potential burden of unnecessary treatment changes and risks for patients.”
The study was funded by AbbVie. Møller-Bisgaard reported receiving grants and nonfinancial support from AbbVie during the conduct of the study and personal fees from Bristol-Myers Squibb outside the submitted work. Multiple coauthors and the editorialists report various financial relationships with multiple companies, the full list of which can be found in the journal. One coauthor (Hetland) is chair of the DANBIO registry, which is located at her institution and which receives funding annually from all providers of biological drugs in Denmark that have an agreement regarding postmarketing data. Kavanaugh has disclosed no relevant financial relationships.