Background Activated synovial fibroblasts (SF) are key mediators of inflammation and joint destruction in rheumatoid arthritis (RA), their activated phenotype has been likened to that of cancer associated fibroblasts. Two major historical limitations to SF study are (a) a paucity of robust in vitro functional assays & (b) comparisons are limited to SF from multi-treated longstanding RA and from osteoarthritis as a non-inflammatory control. We hypothesised that SF migration and wound healing would differ between normal, resolving and persistent disease

Objectives To develop a novel panel of in vitro functional assays to assess SF migration and proliferation, comparing SF in five distinct outcome groups including treatment naïve RA, resolving arthritis and normal synovium

Methods Synovial biopsies were obtained from treatment naïve patients with inflammatory arthritis and SF cultured and expanded to P4-P5 using established methods. Samples taken within 3 months of symptom onset to analyse differences between resolution and persistence were categorised at follow up as either spontaneously resolving arthritis (resolving, n=5) or very early persisting RA (VeRA, n=5). Samples taken from treatment naive established RA of 4-156 months duration (Est. RA (n=5)) and from longstanding multi-treated RA obtained during joint replacement (Est. RA JRep, n=9) were used to analyse the effect of disease duration on RA SF. Normal SF (n=7) were collected during arthroscopy for non-inflammatory knee arthralgia

1. Migration: Cells were seeded in tissue culture migration inserts (Thistle scientific) in 6 well plates. After 24 hour incubation, inserts were removed revealing two cell monolayers separated by a gap. Migration was assessed as percentage of the gap covered in 18 hours. 2. Wound healing scratch test: SF were grown in 6 well plates for 7 days then a standardised gap was created by scratching the monolayer with a pipet tip. Migration was assessed as percentage of gap covered in 18 hours. 3. Proliferation was determined using EDU (Invitrogen) incorporation and flow cytometry. The effect of pro-inflammatory cytokines was assessed by stimulation with TNFaand TGFb

Results Statistically significant differences in pure migration were observed between the five groups (p=0.002). Normal and resolving lines migrated faster than RA lines regardless of TNFa or TGFb stimulation; the biggest difference was between normal/resolving and Est. RA JRep (73.9% vs. 15.9%, p<0.005; 60.8% vs. 15.9%, p=0.05 respectively). VeRA and Est RA samples showed intermediate migration between these groups suggesting decreased migration with increasing disease duration and a degree of functional plasticity in early disease. By contrast, no significant differences were seen in wound healing. There were no differences in EDU incorporation between groups, indicating that the observed effect was not secondary to differential cell proliferation

Conclusions Neither wound healing by migration across pre-cultured surfaces nor cellular proliferation differ between SF from different groups. However de novo migration is high in normal and resolving SF, with attenuation in RA that is greater in longstanding disease. This indicates a novel acquired phenotype that functionally differentiates SF. Studies of cellular migration mechanisms are ongoing

Disclosure of Interest None Declared