Background Better objectives measures for evaluating disease activity and anti-TNFα response in patients with AS are needed. In this sense, MMP-3 seems to be the most promising biomarker although published data are not conclusive (1). Moreover, there is no evidence anti-TNFα therapy inhibit radiological progression, which has brought the concern if inflammation and bone proliferation are or not related processes (2-5). The study of biomarkers implicated in these processes such as dikkopf-1 (DKK-1) and sclerostin will help us to understand it.

Objectives To evaluate the change of MMP-3, DKK-1 and sclerostin serum levels after anti-TNFα therapy and to investigate if any of them are correlated with disease activity or can be useful to predict anti-TNFα response in patients with AS.

Methods From November 2012 to July 2011, patients with AS (New York criteria) who initiated anti-TNFα therapy in a University hospital were included. Socio-demographic and disease characteristics variables were recorded at baseline. Before and after 3 months of therapy, disease activity was measured, and blood samples for determination of serum levels of MMP-3, DKK-1 and sclerostin by enzymoinmunoanalysis were collected. Treatment response was defined using BASDAI50 and ASDAS response. The Wilcoxon test was used for evaluations of changes under treatment and the Mann-Whitney U test for between-group comparisons. Receiver operating characteristic analysis to determine the accuracy to predict response and Correlation testing were performed.

Results A total of 20 AS patients were included, 86% men, 42.4 (31-49) years old, with disease duration 6.8 (3-10) years, 83% HLA-27 positive, from which 80% and 20% received adalimumab and etanercept, respectively. After 3 months of anti-TNFα, all disease activity parameters improved significantly (BASDAI 6.5 vs 3.6; p=0.001, ASDAS 3.6 vs 2.1; p<0.001, CRP 13.5 vs 4.4 mg/L; p=0.001). Serum levels of MMP-3 decreased (100.0 ng/ml vs 68.1; p<0.05) while levels of DKK-1 and sclerostin did not change significantly (7.07 pmol/l vs 7.65; P=0.5 and 21.7 pmol/l vs 22.7, p=0.5, respectively). A significant correlation was only found between baseline levels of MMP-3 and pain VAS. Moreover, we compared biomarker changes based on treatment response, and observed that MMP-3 decreased only in patients who responded to anti-TNF therapy (122.2 ng/ml vs 64.1; p=0.01). Baseline serum levels of MMP-3 were significantly higher in patients who had a good response to anti-TNFα vs those who did not respond (122.9 ng/ml vs 58.9; p<0.05). The area under the curve for MMP-3 to predict BASDAI50 and ASDAS response was 0.73 and 0.78, respectively. The best cut-off was established for levels higher than 59.5 ng/ml, with sensibility of 79-85% and specificity of 50-57%.

Conclusions Serum levels of MMP-3 decreased after anti-TNF therapy and were useful to predict response to this therapy in patients with AS. There was no correlation between serum levels of MMP-3, DKK-1 or sclerostin and disease activity parameters.

1. Sieper J. Curr Opin Rheumatol. 2009 Jul;21(4):335-9

2. Chen CH et al. Rheumatology 2006;45:414–420

3. Wendling D et al. Joint Bone Spine. 2008 Oct;75(5):559-62

4. Arends S et al. J Rheumatol. 2011 Aug;38(8):1644-50

5. Wendling D et al. J Rheumatol. 2007 Aug;34(8):1647-9

Disclosure of Interest None Declared