Background Psoriatic spondyloarthritis (PsSpA), peripheral-only psoriatic arthritis (pPsA) and ankylosing spondylitis (AS) are characterised by differing patterns of osteoproliferation and bone resorption, which may be reflected by levels of serum-soluble bone-turnover biomarkers.
Objectives To investigate four such potential biomarkers (Dkk-1, M-CSF, MMP-3 and OPG) as predictors of: (i) PsSpA occurrence; (ii) axial radiographic severity; (iii) psoriatic arthritis mutilans (PAM) occurrence.
Methods A prospective single-centre cross-sectional study (ADIPSA) recruited pPsA (n=127; fulfilling CASPAR criteria), PsSpA (n=118; psoriasis with radiographic sacroiliitis and/or spondylitis) and AS (n=157; fulfilling modified New York criteria) cases, and assessed them clinically, radiographically, and in terms of serum biomarkers. Serum was obtained from two other centres of clinically characterised psoriasis-only (PsC; n=200) cases, and sex/age-matched healthy controls (HC; n=50). ELISA was used to measure serum Dkk-1, M-CSF, MMP-3 and OPG concentrations. Multivariable regression analyses compared serum concentrations across disease groups.
Results OPG is a biomarker of axial disease in spondyloarthritis (SpA) cases; concentrations were significantly lower in SpA cases with axial disease (RAD) than in those without (non-RAD) (adjusted odds ratio, ORadj 0.20 per ng/ml increase in concentration; 95%CI 0.05, 0.80; p=0.02), independently of having psoriasis.
Dkk-1 is a biomarker of axial disease in SpA cases, with a pattern for increasing concentration along a spectrum of increasing axial involvement. Dkk-1 concentrations were significantly higher in SpA cases with axial disease than in those without (ORadj 1.22; 95%CI 1.05, 1.42; p=0.01), independently of having psoriasis, and significantly lower in PsSpA than AS cases (ORadj 0.85; 95%CI 0.74, 0.98; p=0.02).
M-CSF is a biomarker of arthritis. Compared with HC, M-CSF concentrations were significantly lower in pPsA (ORadj 0.14; 95%CI 0.06, 0.32; p<1x10–5), PsSpA (ORadj 0.07; 95%CI 0.03, 0.17; p<1x10–8), and AS (ORadj 0.37; 95%CI 0.16, 0.85; p<1x10–7). Similarly, M-CSF concentrations were significantly lower in PsA than PsC cases (ORadj 0.44; 95%CI 0.24, 0.82; p=0.01).
MMP-3 is a biomarker of arthritis. Compared with HC, MMP-3 concentrations were significantly higher in pPsA (ORadj 1.06; 95%CI 1.01, 1.10; p=0.02), PsSpA (ORadj 1.06; 95%CI 1.01, 1.11; p=0.02), and AS (ORadj 1.06; 95%CI 1.01, 1.11; p=0.01). Similarly, MMP-3 concentrations were significantly higher in PsA than PsC cases (ORadj 1.02; 95%CI 1.01, 1.03; p=0.0004).
There were no significant associations between biomarkers and either morphology or radiographic severity (mSASSS or PASRI).
Biomarker levels were no different in PsA cases with/without PAM.
Conclusions OPG and Dkk-1 appear to be biomarkers of axial disease in SpA patients (pPsA, PsSpA, AS), independently of cutaneous psoriasis status. M-CSF and MMP-3 appear to be biomarkers of arthritis, but don't differentiate SpA phenotypes. No biomarkers for PAM occurrence in PsA were found. The high concentration of Dkk-1 in AS and PsSpA, supports previous evidence that Dkk-1 is dysfunctional in axial SpA.
Disclosure of Interest None declared