Tartrate-resistant acid phosphatase 5b, but not periostin, is useful for assessing Paget's disease of bone.

Abstract

Periostin is a matricellular protein with a preferential location in cortical bone and periosteal tissue, and tartrate-resistant acid phosphatase 5b (TRAP5b) is a marker of osteoclast numbers. In Paget's disease of bone (PDB), there is increased cortical thickening and probably increased periosteal apposition, along with increased osteoclast numbers. To analyse if circulating periostin is a biomarker for PDB, and if it is associated with disease activity and involvement of long bones that represent major cortical contribution. Also, to analyse whether TRAP5b, a scarcely explored bone resorption marker, is useful in the assessment of PDB. We recruited 42 patients with PDB (13F/29M; 71 ± 11.6 yrs). 71.4% had active disease, 66.6% had polyostotic disease and 54.8% had long bone involvement. Blood and urine samples were taken between 8:00 and 10:00 A.M. after an overnight fast. Periostin and TRAP5b were measured in serum, using commercial ELISA assays (Biomedica and IDS, respectively). Serum total ALP, PINP, CTX, bone ALP and urinary NTX were measured. Reference values for periostin and TRAP5b were obtained from 45 healthy subjects. Serum periostin did not differ between patients and controls (989.4 ± 173.2 vs. 966.9 ± 195.4 pMol/L, p = 0.572). No significant differences were observed between patients with and without active disease (964.5 ± 168.8 vs.1051.6 ± 175.6 pMol/L, p = 0.143), involvement or not of long bones (1022.2 ± 145.8 vs 949.7 ± 198.2 pMol/L, p = 0.181) and monostotic or polyostotic disease (963.8 ± 198.7 vs 1002.2 ± 161.4 pMol/L, p = 0.505). There were significant correlations between serum periostin and all bone turnover markers (bone ALP, PINP, uNTX, sCTX and TRAP5b) in PDB patients with active disease, but not in the inactive PDB group. Serum TRAP5b was significantly higher in PDB patients than in controls (4.43 ± 1.76 vs. 3.21 ± 1.02 U/L, p < 0.001), in those with active disease (4.98 ± 1.76 vs. 3.07 ± 0.72 U/L, p < 0.001) and in patients with polyostotic disease than in those with monostotic disease (4.81 ± 1.79 vs 3.68 ± 1.5 U/L, p = 0.005). TRAP5b levels were not influenced by previous bisphosphonate treatment (4.14 ± 1.42 vs. 4.84 ± 2.02 U/L, p = 0.206). Periostin is not useful for assessing PDB, whilst TRAP5b, which has been a scarcely explored bone turnover marker until now, may be useful in the analysis of this disease, providing new information on the resorption process. In addition, periostin levels correlate with all classical BTMs in active PDB, suggesting that this marker may reflect periosteal and cortical metabolism in accelerated bone turnover states.