Background/Purpose: Periostin is a secreted, homodimeric protein that is synthesized by mesenchymal cells of the periosteum, which is involved in osteoblast homeostasis of the periosteum and could be one of the mediators of local bone formation in response to mechanical stress and/or inflammation. Our hypothesis is that periostin could be involved in the syndesmophyte formation in spondyloarthritis. The objective was to assess the relationships between serum periostin levels, bone formation regulators (sclerostin, Dickkof-1 (DKK-1)) and presence of syndesmophyte, in a cohort of early inflammatory back pain (IBP) suggestive of axial spondyloarthritis.  

Methods: The DESIR cohort is a prospective, multicenter French cohort of patients with early IBP (Calin or Berlin criteria) (>3 months and <3 years of duration) suggestive of axial SpA, including 708 patients. All SpA patients were naive of any TNF blocker at inclusion in the study. Serum periostin levels were assessed at baseline in the whole cohort by sandwich enzyme-linked immunosorbent assay (ELISA) (Nordic Biosite, Sweden). Demographic characteristics, disease activity parameters, bone mineral density measurements were assessed at baseline and 2 years.  DKK-1 and SOST serum levels (by sandwich ELISA), radiographs and MRI of axial skeleton (spine and sacroiliac joints) were performed at baseline. Correlations between serum periostin levels with baseline and 2 year variables were tested. 

Results: Serum periostin was available in 701 patients (mean age 33.7±8.6 years, 46.2% of males). Mean (±SD) serum level of periostin was 396.5±461.4 pg/mL. Levels were higher in men (431.5±486.5 pg/mL) than in women (366.4 ±113.7 pg/mL) and decreased with age. Patients with at least one syndesmophyte at baseline had a lowest level of serum periostin than patients without syndesmophytes (222.4 ±286.5 ng/ml vs 412.89 ±471.3 ng/ml, respectively, p=0.021). There was a significant correlation between serum periostin levels and baseline sclerostin levels (r= -0.09; p=0.02) and baseline DKK-1 levels (r=-0.12; p=0.002). Serum periostin levels were weakly correlated with 2-year changes in disease activity parameters: CRP (p=-0.1; 0.027), ESR (r=-0.14; p=0.002), BASDAI (r=-0.12; p= 0.003) and ASDAS CRP (r=-0.18; p=≤0.0001). We did not find any difference between patients with and without the presence of bone marrow oedema on MRI. Serum periostin was not correlated with low BMD (Z≤-2 at at least one site) and was not a predictor of significant bone loss (decreased in BMD ≥0.03g/cm at at least one site).

Conclusion: This study conducted in a large cohort of patients with early axial SpA shows that low serum level of periostin is associated with local bone formation. Periostin is correlated to serum sclerostin and DKK-1 levels and parameters of inflammation. Relationship between serum periostin and formation of new syndesmophyte will be studied using the 2-year centralized analysis of X-rays and MRI.

---

« Back to 2013 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/serum-periostin-a-new-marker-of-local-bone-formation-in-early-inflammatory-back-pain-results-from-the-desir-cohort/