Background/Purpose:  Rheumatoid arthritis is associated with an excess cardiovascular disease (CVD) burden. Fewer CVD events associated with specific DMARD therapies have been reported; however, causal effects on atherosclerosis have never been evaluated rigorously in a randomized trial. Radiolabeled fluorodeoxyglucose (FDG) is taken up by macrophages in atherosclerotic plaques, is detectable with positron emission tomography (PET), and is a surrogate for atherosclerosis. We designed a multi-center randomized controlled clinical trial to: 1) Compare the change (over 24 weeks) in FDG uptake in the aorta and carotid arteries among methotrexate (MTX) inadequate responders who will be randomized to adding either a TNF inhibitor (i.e. step-up MTX + TNFi) or sulfasalazine + hydroxychloroquine (i.e. step-up triple therapy), and 2) assess whether achieving low RA disease activity or remission (LDAR) is associated with a parallel reduction in arterial FDG uptake. Correlations of change in arterial FDG uptake with change in circulating biomarkers will also be performed.

Methods:  The primary hypothesis of the Treatments Against Rheumatoid Arthritis and Effects on FDG-PET (TARGET) Trial is that TNFi +MTX will reduce vascular inflammation to a greater extent than triple therapy after 24 weeks. Secondarily, RA patients with LDAR at 24 weeks will have less vascular inflammation than those with persistent moderate/high disease activity. The primary outcome measure will be the maximal aortic FDG uptake of the most diseased segment (MDSmax) standardized to the blood pool in the superior vena cava [i.e. the target to background ratio (TBR)]. Sample size projections are based on estimates of change in aortic FDG uptake from preliminary sources, 0-15% dropout, 0-15% unintended crossover into the TNFi arm, a conservative 30% LDAR achieved, at least 90% power for the primary and secondary analyses, a two-tailed alpha=0.05, and equal allocation to the treatment arms.

Results:  From a published study of 17 RA patients, the mean ± SD baseline aortic MDSmax TBR was 2.51±0.33 units. In the same study, there was a 0.46 unit reduction in this measure after 8 weeks of TNFi. We calculated that the trial would require a pooled sample size of n=170 (n=85 per arm) to detect an absolute difference in the 6-month MDSmax TBR of 0.17 units between the treatment groups with 90% power should the highest projected crossover occur. This same sample size would also have 90% power to detect a 0.19 unit difference in MDSmax TBR between those achieving LDAR vs. moderate or high disease activity, using the same assumptions. The magnitude of these differences is consistent with studies comparing low vs. high dose statin use. Conservatively assuming a 15% drop-out, 100 patients per arm will be enrolled beginning in July 2016.

Conclusion: The TARGET trial, the first to explore the effect of immunomodulation on vascular inflammation in any rheumatic disease, is powered to detect meaningful differences in vascular inflammation between treatments. Further, the minimal detectable difference in arterial inflammation between those achieving LDAR compared with those not responding as may be small, yet clinically meaningful.

« Back to 2016 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/delineating-the-effects-of-rheumatoid-arthritis-pharmacotherapies-on-vascular-inflammation-rationale-and-design-of-a-clinical-trial/