Background/Purpose: The Multi-Biomarker Disease Activity (MBDA) score was developed to provide an objective measure of rheumatoid arthritis (RA) disease activity. Demonstrating moderate correlations with conventional measures of RA disease activity, the capacity of the MBDA to predict treatment response may vary among available disease-modifying agents. The performance of the MBDA in patients initiating methotrexate (MTX), a cornerstone therapy in RA, has not been evaluated to date.

Methods: We conducted a secondary analysis of a 16-week, open-label study of RA patients initiating MTX (15mg/wk) with dose escalation to 20mg/wk for those failing to achieve disease remission (DAS28 <2.6) at 8 weeks. The primary study outcome was the proportion achieving ACR50 response at week 16. MBDA scores were measured using banked serum collected at baseline (N=130) and week 16 (N=95). The association of baseline MBDA with treatment response was assessed using multivariable regression models adjusting for age, sex, race, smoking status, and rheumatoid factor (RF) concentration. Convergent validity and responsiveness were determined by calculating correlations of the MBDA with DAS28-ESR, ESR, patient global assessment (PtGA), Health Assessment Questionnaire (HAQ), as well as correlations of MBDA change with changes in alternative disease activity measures and through the calculation of standardized response means (SRM).

Results: Patients had a mean age of 54 yrs and were predominantly female (75%), seropositive (73% for RF, 75% for anti-CCP), and had high baseline disease activity (DAS28-ESR of 5.3, SD 1.2). Mean baseline MBDA score was 48.3 (SD 17.2) with 12%, 28%, and 61% categorized as low, moderate, and high disease activity by MBDA. After 16 weeks, the MBDA improved by a mean of 9.4 (SD 14.5) and DAS28-ESR improved by a mean of 1.9 (SD 1.5). Patients achieving an ACR50 response demonstrated greater reductions in MBDA than those failing to achieve an ACR50 response (p=0.01). However, baseline MBDA scores (OR 1.01, 95% CI 0.99, 1.04) and categories (moderate: OR 1.55, 95% CI 0.45-14.20; high: OR 2.55, 95% CI 0.50-12.98) were not predictive of response. Higher baseline MBDA scores were associated with greater improvement in DAS28-ESR (p=0.01), as were baseline DAS28-ESR values (p<0.001). The MBDA demonstrated moderate correlations with DAS28-ESR and ESR, but weaker correlations with the HAQ and PtGA (Table 1). Treatment responsiveness was greater for DAS28-ESR (SRM -1.32) than MBDA (SRM -0.65).

Conclusion: In this 16-week open-label study of MTX in RA patients, baseline MBDA was associated with greater reductions in DAS28-ESR but was not a robust predictor of ACR50 response. Although demonstrating moderate convergent validity with standard disease activity assessments, the MBDA yielded lower responsiveness than the DAS28-ESR in the context of MTX treatment.

« Back to 2018 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/assessment-of-treatment-response-using-the-multi-biomarker-disease-activity-score-in-rheumatoid-arthritis-patients-initiating-methotrexate/