Background Studies conducted prior to the widespread usage of biologic therapy in patients with rheumatoid arthritis (RA) can provide valuable information regarding the natural course of the disease and the potential prognostic value of biomarkers. The Leiden Early Arthritis Clinic (EAC) cohort comprises a group of subjects with recent onset arthritis who were enrolled at the Department of Rheumatology of Leiden University Medical Center beginning in 1993. A novel multi-biomarker test for disease activity (MBDA) in RA has recently been developed and validated, and it was of interest to assess the relationship between MBDA and radiographic progression in this well-characterized cohort.
Methods Subjects were members of the Leiden EAC cohort and fulfilled the American College of Rheumatology (ACR) 1987 revised criteria for the classification of RA (n = 163 unique subjects, with 271 total visits). Therapy consisted mainly of DMARDs, with less than 5% of subjects receiving anti-TNFs. The pre-specified MBDA algorithm integrated the concentrations of 12 biomarkers to produce scores from 1-100: VCAM-1, EGF, VEGF-A, IL-6, TNF-RI, YKL-40, MMP-1, MMP-3, leptin, resistin, SAA, CRP. Each study visit included assessment of the MBDA score and evaluation of the Sharp van der Heijde score (SHS), and was followed by calculation of a second SHS score 12 months later. Radiographic progression was assessed by calculating the change in SHS over 12 months. The association between MBDA score and rapid progression (ΔSHS > 5) was assessed using the area under the receiver operating characteristic curve (AUROC). The profile of radiographic progression as a function of MBDA score was studied using quantile regression, a robust method, which is an extension of mean regression that can provide more information about rapid progressors. The effect of anti-CCP status on this relationship was also explored.
Results Mean age was 55, 67% of the cohort was female, 66% was anti-CCP-positive, and the median SHS score was 23. The risk of radiographic progression increased with increasing MBDA scores (AUROC = 0.77 for differentiating rapid progressors, n = 271, p < 0.001). The fitted quantile regression curves (Graph 1) also imply that increases in the risk of progression are associated with increases in the MBDA score. As shown in the graph, the rapid progressors were observed to have the highest rate of increase in the risk of progression, and the rate varied across the range of MBDA scores for all types of progressors, with the most prominent increase occurring between the scores of 40 and 60.
Conclusions In the Leiden early arthritis cohort, the biomarker-based MBDA score was associated with risk of radiographic progression over the ensuing 12 months, and risk increased nonlinearly as a function of the score. Identification of patients at risk of rapid radiographic progression may be aided by assessing the MBDA score.
Disclosure of Interest W. Li Shareholder of: Crescendo Bioscience, Inc., Employee of: Crescendo Bioscience, Inc., A. van der Helm-van Mil Grant/research support from: The Netherlands Organization for Health Research and Development, R. Knevel Grant/research support from: Dutch Arthritis Institute, G. Cavet Shareholder of: Crescendo Bioscience, Inc., Consultant for: Crescendo Bioscience, Inc., T. Huizinga: None Declared, D. Haney Shareholder of: Crescendo Bioscience, Inc., Employee of: Crescendo Bioscience, Inc.