In previous works, we demonstrated that markers of systemic inflammation (elevated C-reactive protein (CRP) and erythrocyte sedimentation rate (ESR)) are independently associated with radiographic spinal progression over 2 years in axial spondyloarthritis (axSpA) in the German spondyloarthritis inception cohort (GESPIC).1 We also found previously that calprotectin, which is secreted during monocyte infiltration into inflamed tissues, and thus, directly reflects a potentially important pathophysiological mechanism in SpA,2 ,3 and which was found to be associated with structural joint damage in psoriatic arthritis4 and in rheumatoid arthritis,5 is clearly elevated in SpA as compared to healthy controls, and decreases rapidly and consistently upon effective treatment,2 although in another work, no differences in serum calprotectin between ankylosing spondylitis patients and healthy controls could be found.6 We aimed here to assess whether serum calprotectin levels are predictive for progression of structural damage in the spine in axSpA.

Seventy-six patients (mean age 38.0±11.5 years, mean symptom duration 4.6±2.8 years, 66% males, 82% HLA-B27-positives) with definite axSpA (n=63 fulfilling the modified New York criteria for ankylosing spondylitis and n=13 patients with axSpA not fulfilling the radiographic part of the modified New York criteria, but with at least one syndesmophyte in the cervical or lumbar spine) from GESPIC,1 ,7 were selected for this analysis. Detailed description of the cohort and radiographs scoring has been reported elsewhere.1 ,7–9 Radiographic spinal progression was defined as (1) worsening of the modified Stoke Ankylosing Spondylitis Spine Score (mSASSS) by ≥2 units, and (2) development of a new syndesmophyte or progression of existing syndesmophytes (formation of a bridging syndesmophyte) after 2 years. Levels of serum calprotectin were determined by ELISA (Hycult Biotech, the Netherlands).

Baseline calprotectin serum level was significantly higher in patients with mSASSS worsening (n=15) versus those without mSASSS worsening (n=61): mean±SD 0.68±0.21, median 0.69 μg/mL vs mean±SD 0.48±0.26, median 0.42 μg/mL, respectively, p=0.005 (figure 1A); as well as in patients with syndesmophyte formation / progression (n=18, 12 of them fulfilled also the mSASSS worsening criterion) versus those without (n=58): mean±SD 0.64±0.27, median: 0.65 μg/mL vs mean±SD 0.48±0.25, median 0.42 μg/mL, respectively, p=0.035 (figure 1B).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Calprotectin serum levels in patient with and without radiographic spinal progression defined as an mSASSS worsening by ≥2 points over 2 years (A) and syndesmophyte formation/progression (B). Each box indicates the median value, the 1st and the 3rd quartiles; whiskers demonstrate minimal and maximal values, circles—outliers, crosses inside the boxes—mean values. p Values refer to the independent samples t test. mSASSS, modified Stoke Ankylosing Spondylitis Spine Score.

A receiver operating characteristic (ROC)-analysis showed a good performance of calprotectin in prediction of mSASSS worsening (area under the curve (AUC)=0.740 (95% CI 0.614 to 0.866); p=0.004; figure 2A), as well as in prediction of syndesmophyte formation/progression (AUC=0.670 (95% CI 0.520 to 0.819); p=0.031; figure 2B). In patients with axSpA who already had syndesmophytes at baseline (n=49; 65%), calprotectin demonstrated similar performance, compared to the whole group, as a predictor of radiographic spinal progression: AUC=0.741 (95% CI 0.597 to 0.884), p=0.009 for mSASSS worsening and AUC=0.706 (95% CI 0.547 to 0.866), p=0.020 for syndesmophyte formation / progression.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2

Receiver operating characteristic analysis of the association between baseline serum level of calprotectin and radiographic spinal progression defined as an modified Stoke Ankylosing Spondylitis Spine Score (mSASSS) worsening by ≥2 points over 2 years (A) and syndesmophyte formation/progression (B) in patients with axial SpA (n=76).

Calprotectin serum level >0.5 μg/mL (the optimal predictive cut-off in the ROC-analysis) had a sensitivity of 80%, a specificity of 62%, and an OR=6.2 (95% CI 1.6 to 24.2), p=0.009, as a predictor of mSASSS worsening. As a predictor of syndesmophyte formation/progression, calprotectin serum level >0.5 μg/mL had a sensitivity of 72%, a specificity of 60%, and an OR=4.0 (95% CI 1.2 to 12.6), p=0.020.

Although calprotectin serum level demonstrated a positive correlation with acute-phase reactants (CRP: Spearman's r=0.382, p=0.001, ESR: r=0.401, p<0.001) and showed an overall performance similar to CRP in prediction of radiographic progression, the association of high calprotectin with radiographic spinal progression remained significant after adjustment for CRP and other risk factors (syndesmophytes at baseline and smoking): OR=5.5 (95% CI 1.2 to 25.8), p=0.030, for mSASSS worsening and an OR=4.1 (95% CI 1.2 to 32.0), p=0.028 for syndesmophyte formation/progression.

Also, calprotectin demonstrated a significant correlation with vascular endothelial growth factor (VEGF)—another biomarker with predictive value for radiographic spinal progression in axSpA9: Spearman's r=0.552, p<0.001. After adjustment for VEGF, calprotectin remained significantly associated with mSASSS worsening (OR=4.7 (95% CI 1.1 to 20.5), p=0.040) while the association with syndesmophyte formation has lost statistical significance (OR=3.6 (95% CI 1.0 to 13.7), p=0.056).

Thus, calprotectin represents a novel predictive biomarker for radiographic spinal progression in axial SpA.