Background We previously reported increased expression of cell membrane RANKL in PBMC of patients with active rheumathoid arthritis (RA) which was down-regulated by anti TNF-α treatment with adalimumab, while soluble RANKL and OPG were scarcely affected. We speculated that anti TNF-α induced down-regulation of membrane RANKL could be important in preventing articular damage in RA patients. Joint damage may also be mediated by the balance of other mediators involved in osteoclast functions such as LIGHT (TNFSF14, member of the TNF superfamily) and cathepsin-K, while osteoblast functions are influenced by DKK-1 and sclerostin, although the data available is still contrasting and the significance under debate.
Objectives To investigate the impact of anti TNF-α treatment on the major soluble mediators involved in bone homeostasis in RA patients.
Methods The effects of anti TNF-α therapy on bone homeostasis was studied in 15 active RA patients (DAS28 5.9±0.9) and compared with 20 healthy controls (HC); data were collected at baseline, after 6, 12 and 24 weeks. Adalimumab 40 mg was administered every other week according to guidelines, all patients experienced a satisfactory clinical response according to EULAR criteria. The serum levels of DKK-1, Sclerostin, Cathepsin-K were measured by enzyme-linked immunoassorbent assay (ELISA), all purchased from Biomedica (Vienna, Austria); LIGHT protein was detected with Quantikine ELISA (R&D System, Europe, UK) according to the manufacturer’s instruction. Values are presented as the median and interquartile range. Serum levels at different times after treatments were compared with those before therapy and HC. The significance of the results was analysed using the non parametric Mann-Whytney U-test or Wilcoxon, as appropriate, using Prism 5.0 software (GraphPadInc). P values less than 0.05 were considered significant.
Results Cathepsin-K levels were found to be higher in RA patients compared to HC (9.1 IQR 45-14.3 pmol/mL) both before treatment with adalimumab (12.7 IQR 11.3-15.3 p=0.019) and at W6 (15.5 IQR 12.5-26.8 p=0.001), W12 (16.4 IQR 12.1-20.5 p=0.003), W24 (15.0 IQR 12.9-16, p=0.003). LIGHT levels were also consistently higher in the RA group compared with HC (66.0 IQR 42.6-112.0 pg/L) at baseline (96,6 IQR 65.3-179.2 p=0.03), W6 (101.6 IQR 88.7-203.1 p=0.006), W12 (150.4 IQR 88.8-192.2 p=0.003) and W24 (138.9 IQR 66-228.5 p=0.015). Sclerostin levels were higher in the untreated RA group (31.5 IQR 26.3-48.7 pmol/L) compared to HC (24.5 IQR 18.7-29.6 p=0.007). It is noteworthy that long term adalimumab treatment induced a significant reduction in sclerostin levels at W24 (19.50 IQR 14.6-29.9 p=0.002); conversely, no significant changes in DKK-1 levels were observed following adalimumab treatment at baseline (21.1 IQR 7.9-57pmol/L) compared to W6 (19.5 IQR 6.7-59.5), W12 (20.1 IQR 9.2-57.5), W24 (11.1 IQR 5.2-54.6).
Conclusions The decreased levels of sclerostin following anti TNF-α treatment may reduce the inhibition of theWNT/β-catenin pathway leading to increased osteoblast activity which may balance the increased osteoclast function in RA patients (as mirrored by the persistent increase in cathepsin-K and LIGHT) and therefore contributing to the inhibition of joint damage seen in patients treated with anti-TNF-α drugs.
Disclosure of Interest None Declared