Background Biological disease-modifying antirheumatic drugs (bDMARDs) inhibit progression of structural damage in rheumatoid arthritis (RA). These results suggest the possibility that bDMARDs improve osteoclastic bone destruction of RA. However, the detailed mechanism of bDMARDs for bone metabolism in RA is poorly understood.
Objectives To clarify the mechanism of tocilizumab (TCZ) or abatacept (ABT) for bone metabolism in active RA.
Methods We selected 80 female patients with active RA, 44 patients were treated with TCZ and 36 patients were treated with ABT intravenously. Next, Circulating levels of type I collagen cross-linked N-telopeptides (NTx), osteocalcin (OC), soluble receptor activator of NF-kappa B ligand (sRANKL), osteoprotegerin (OPG), Dicckopf-1 (DKK-1), and osteopontin (OPN) were examined by ELISA at baseline and after 12 weeks of each treatment.
Results Matching of patients according to propensity score resulted in a cohort that consisted of 28 patients in TCZ group and 28 patients in ABT group. Patient's background between TCZ group and ABT group, including age, prednisolone or methotrexate dose, and baseline of DAS-28 was matched. In TCZ group, average of NTx, DKK-1 and OPN levels at 12 weeks decreased significantly from the baseline (24.4 vs 21.5 nmol BCE/L; p<0.05, 2743 vs 2138 pg/mL; p<0.01, 90 vs 60 pg/mL; p<0.01 respectively). Average of OC levels at 12 weeks increased significantly from the baseline (8.6 vs 10.1 ng/mL; p<0.01). Average of sRANKL and sRANKL/OPG levels at 12 weeks tended to decrease from the baseline (0.47 vs 0.41 pmol/L, 0.13 vs 0.10% respectively). Interestingly, average of OPG levels at 12 weeks tended to increase from the baseline (5.03 vs 5.23 pmol/L). In ABT group, similarly, average of NTx and OPN levels at 12 weeks decreased significantly from the baseline (16.1 vs 14.7 nmol BCE/L; p<0.05, 86 vs 71 pg/mL; p<0.05 respectively). OC levels tended to increase (6.4 vs 6.8 ng/mL). However, sRANKL, OPG and sRANKL/OPG levels were not changed. In contrast, average of DKK-1 levels at 12 weeks increased significantly from the baseline (2336 vs 2558 pg/mL; p<0.05). In comparison of the rate of change from the baseline (%change) of these biomarkers between TCZ group and ABT group, %change of OPG in TCZ group increased significantly compared with ABT group (5.56 vs -1.77%; p<0.05) and %change of DKK-1 in TCZ group decreased significantly compared with ABT group (-18.2 vs 10.4%; p<0.01).
Conclusions TCZ or ABT has improved inflammatory bone destruction of RA. However, the main mechanism of TCZ and ABT is different. These results suggest that TCZ has improved bone metabolism in RA through the control of osteoclastogeneis via RANKL/OPG balance. Especially, the promotion of osteoblastogenesis via the inhibition of DKK-1 may be a specific effect in TCZ compared with ABT. On the other hands, In ABT, it is suggested that the suppression mechanism involved in osteoclastogenesis which does not pass control of RANKL/OPG balance exists. This mechanism may be a direct inhibiting effect of osteoclast precursors.
Disclosure of Interest None declared