Background/Purpose: To determine whether a multi-biomarker disease activity (MBDA) score could predict radiographic damage progression in patients with rheumatoid arthritis (RA).

Methods: The BeSt study enrolled 508 patients. Of these, 84 patients had MBDA scores at baseline and radiographs of hands and feet at baseline and 1 year later and 81 patients had MBDA scores at 1 year and radiographs at 1 year and at 2 years. Radiographs were scored by two independent blinded readers using the Sharp van der Heijde Score (SHS).

MBDA scores were calculated using a validated algorithm (Vectra® DA algorithm score) integrating the levels of 12 biomarkers, (VCAM-1, EGF, VEGF-A, IL-6, TNF-RI, MMP-1, MMP-3, YKL-40, leptin, resistin, SAA, CRP) for assessing disease activity in patients with RA. The generated MBDA score ranges between 1 and 100, and can be categorized in low (≤ 29), moderate (>29 – 44) and high (>44) disease activity levels.

Receiver Operating Curves (ROC) were used to calculate an area under the curve (AUC) with outcome any radiographic damage progression (increase ≤ 0.5 points SHS) yes or no. Poisson regression with increase in SHS as continuous outcome was used to calculate a relative risk (RR). In all analyses MBDA scores are compared to disease activity scores (DAS) in predicting radiographic damage in the following 12 months.

Results: Baseline characteristics of patients in this analysis were similar to the entire BeSt cohort. Mean age was 53 and 75% were female. Mean Disease Activity Score (DAS) was 4.30, 62% were rheumatoid factor positive and 56% anti-citrullinated protein antibodies (ACPA) positive.

At baseline, DAS had an AUC of 0.373 (95% CI 0.248 – 0.498) and MBDA had an AUC of 0.606 (95% CI 0.482 – 0.729) on radiographic progression in the subsequent year. At one year, the AUC for DAS was 0.527 (95% 0.392 – 0.729) compared to an AUC of 0.686 (95% CI 0.564 – 0.809) for MBDA.

Poisson regression showed a RR of 1.039 for MBDA at baseline to develop radiographic progression, adjusted for DAS and ACPA, indicating a 1.47 times higher risk of radiologic progression with each 10 points MBDA increase. At 1 year, the RR of MBDA, also adjusted for DAS and ACPA, is 1.037, indicating a 1.44 times higher risk with each 10 points MBDA increase. A high baseline MBDA shows a RR of 3.7 for radiologic progression in the next year compared to low and moderate (≤ 44 ) baseline MBDA levels combined. At 1 year, high MBDA showed a RR of 4.6 compared to low MBDA. Moderate MBDA compared to low MBDA at 1 year does not show a significantly greater risk of radiologic progression at 2 years (Table 1).

Table 1: Poisson regression analysis with MBDA categories and  progression in the subsequent 12 month period.

	Risk for SHS progression, baseline to 1 year
Baseline (N=84)	RR	95% CI	P value
MBDA ≤ 44	ref	ref	ref
MBDA > 44	3.74	1.45 – 9.66	0.006
At 1 year (N=81)	Risk for SHS progression, year 1 to year 2
MBDA ≤ 29	ref	ref	ref
MBDA > 29 – 44	1.44	0.45 – 4.55	0.537
MBDA > 44	4.62	1.34 – 15.95	0.015

Conclusion: Adjusted for DAS and ACPA, baseline MBDA scores predicted radiographic damage progression at 1 year as well as MBDA at 1 year predicted radiographic progression at year 2 during disease course. Therefore, MBDA may be of value in future treatment strategies in RA patients.

---

« Back to 2013 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/a-multi-biomarker-disease-activity-score-for-rheumatoid-arthritis-predicts-radiographic-damage-in-the-best-study/