ANTI-FIBROTICS TO PREVENT NEW BONE FORMATION IN SPONDYLOARTHRITIS: PROOF OF CONCEPT USING PIRFENIDONE IN CELL CULTURE MODELS

Background: The pathogenesis of spondyloarthritis (SpA) involves both inflammation and new bone formation in the spine. In line with this, the disease has been characterized as both inflammatory and fibrotic. Current treatment including inhibitors of tumour necrosis factor alpha (TNFα) seem to dampen inflammation while new bone formation can progress. Therefore, there is an unmet therapeutic need for the treatment of new bone formation in SpA. Fibrosis is mediated by myofibroblasts and new bone formation is the result of increased osteoblast mineralization and decreased osteoclast bone degradation. Here, we evaluate the potential effect of the newly approved anti-fibrotic agent pirfenidone (Esbriet, Pirespa) on fibrosis and new bone formation in cell culture models of SpA. Objectives: We hypothesized that pirfenidone inhibits SpA myofibroblast formation and activity and osteoblast mineralization. Methods: Synovial fluid mononuclear cells from patients with SpA (n=6) were included for culturing fibroblast-like synovial cells (FLSs) while osteoblasts were purchased. The cells were cultured with pirfenidone in increasing concentrations (0.25, 0.5, and 1.0 mg/ml) with or without stimulation with tumor necrosis factor alpha (TNFα), transforming growth factor beta (TGFβ), or interferon gamma (INFγ). The proliferation of FLSs was analyzed with light microscopy and flow cytometry using the marker Ki67. The differentiation and activation of FLSs was assessed with flow cytometry, a proteome profiler assay and enzyme-linked immunosorbant assays. The mineralization capacity of the osteoblasts was measured as deposition of hydroxyapatite. Results: Pirfenidone reduced the Ki67 expression 7.1-fold in untreated FLSs (p=0.001) and 11.0-fold in FLSs stimulated with TGFβ, TNFα, and INFγ (p=0.022). PFD further inhibited TGFβ induced upregulation of αSMA (Figure 2A) and INFγ induced upregulation of HLA-DR (Figure 2C) in all cultures. There was no difference between the percentage of ICAM-1 positive FLSs in cultures treated with or without pirfenidone. In supernatants from FLSs stimulated with TGFβ, TNFα, and INFγ a total of 12 cytokines or chemokines had values above the detection limit in the membrane-based antibody array. Pirfenidone decreased the secretion of 3 of these 12 cytokines or chemokines more than 2-fold. The changes in secretion of monocyte chemoattractant protein 1 (MCP-1) and chitinase-3-like protein 1 (CHI3L1, also known as YKL-40) were validated with ELISA. Further, pirfenidone decreased the secretion of both DKK1 (p=0.006) and OPG (p=0.02) by SpA FLSs stimulated with TGFβ, TNFα, and INFγ, while the concentration of RANKL was below the detection limit of the ELISA assay in all cultures. Finally, pirfenidone inhibited the deposition of hydroxyapatite by osteoblasts in a dose-dependent manner (p=0.0001). This inhibition was partly reversible when removing pirfenidone after the first week of the mineralization assay. Conclusions: Taken together, pirfenidone inhibited SpA myofibroblast formation and activity and osteoblast mineralization. This encourages further research in anti-fibrotics as treatment of new bone formation in SpA. Disclosure of Interest: None declared DOI: 10.1136/annrheumdis-2017-eular.1503Citation: Annals of the Rheumatic Diseases, volume 76, supplement 2, year 2017, page 1088Session: Spondyloarthritis - etiology, pathogenesis and animal models (Abstracts Accepted for Publication )