Arthritis & Rheumatism, Volume 62,
November 2010 Abstract Supplement
Abstracts of the American College of
Rheumatology/Association of Rheumatology Health Professionals
Annual Scientific Meeting
Atlanta, Georgia November 6-11, 2010.
Validation of a Multi-Biomarker Test for Rheumatoid Arthritis (RA) Disease Activity (Vectra DA) in a Multi-Cohort Study.
Curtis9, Jeffrey R., Haney3, Doug, van der Helm5, Annette H. M., Shen3, Yijing, Knevel5, Rachel, Cavet3, Guy, Dirven5, Linda
University of Toronto, Toronto, ON, Canada
Brigham & Womens Hospital, Boston, MA
Crescendo Bioscience, Inc.
Leiden Univ Med Ctr, Leiden, The Netherlands
Leiden Univ Med Ctr
Oklahoma Med Research Foundation, Oklahoma City, OK
Rheumatology Associates, Dallas, TX
Rheumatology Associates of LI, Dix Hills, NY
University of Alabama-Birmingham, Birmingham, AL
Regular measurement of disease activity enables more effective management of rheumatoid arthritis (RA). DAS28CRP is a validated tool for assessment of disease activity, but it requires enumeration of tender and swollen joints, which are subject to significant inter- and intra-assessor variability, and is not routinely performed by all physicians. A test based on serum protein biomarkers involved in RA pathophysiology has the potential to provide quantitative, objective and reproducible information that has not previously been available to clinicians. A 12-biomarker test of RA Disease Activity (DA Test) has recently been developed in a series of clinical studies. Here we show in a prospectively-designed study that the biomarker-based DA Test validly quantifies RA disease activity as compared to the DAS28CRP.
230 RF+ and/or anti-CCP+ patients were selected from participants in the Index for Rheumatoid Arthritis Measurement (InFoRM), Brigham and Women's Hospital Rheumatoid Arthritis Sequential Study (BRASS), and Leiden Early Arthritis Clinic studies. To ensure test validity across a wide range of disease activity, patients representing all levels of disease activity were selected. The concentrations of 12 serum biomarkers (IL-6, EGF, VEGF-A, Leptin, SAA, CRP, VCAM-1, MMP-1, MMP-3, Resistin, YKL-40, and TNF-RI) were measured and combined in a pre-specified algorithm to generate DA Test scores ranging from 1100. The DA Test score was calculated from an equation similar to that for DAS28CRP where scores for TJC28, SJC28, and Patient Global were individually modeled using serum biomarkers. To calculate area under the ROC curve (AUROC), a DAS28CRP cutoff of 2.67[i] was used to assign cases to low vs moderate/high disease activity. A separate, longitudinal analysis of patients from the InFoRM study with significant changes in DAS28CRP was undertaken to understand whether changes in DA Test score were associated with changes in DAS28CRP.
The AUROC for DA relative to the DAS28CRP cutpoint of 2.67 was 0.77 (CI: 0.70 0.83; p < 0.001); the correlation between DA and DAS28CRP was 0.56 (CI: 0.46 0.64; p < 0.001). Decomposition of the DA Test into a non-CRP DA Test score and a CRP term demonstrated that the non-CRP DA Test score was an independent predictor of DAS28CRP (p < 0.001) in multivariate analysis (after controlling for CRP). In longitudinal analysis, there was a significant association between change in DA Test score and change in DAS28CRP (p<0.01). Exploratory analysis suggests the DA Test may more accurately detect low disease activity than CRP (Figure).
Fig 1. Proportion of patients with joint counts below a threshold for those with low CRP or DA test score
The DA Test has been validated as a quantitative, objective, biomarker-based test of RA disease activity. It could be used to optimize clinical care and provide an objective measure of treatment response in observational analyses or randomized trials.
%[i] Inoue et al. Ann Rheum Dis. 2007; 66:407409
To cite this abstract, please use the following information:
Curtis, Jeffrey R., Haney, Doug, van der Helm, Annette H. M., Shen, Yijing, Knevel, Rachel, Cavet, Guy, et al; Validation of a Multi-Biomarker Test for Rheumatoid Arthritis (RA) Disease Activity (Vectra DA) in a Multi-Cohort Study. [abstract]. Arthritis Rheum 2010;62 Suppl 10 :1782