associates with the presence of anti-citrullinated protein antibodies, but not with the onset of arthritis: studies in an at-risk population.

Objective: The antibody response against () is elevated in rheumatoid arthritis (RA), especially in patients with anti-citrullinated protein antibodies (ACPA). Here, we investigated whether antibodies against the virulence factor arginine gingipain (Rgp) are associated with the RA-risk phase and development of arthritis.

Methods: At-risk individuals were included in a prospective study (Risk-RA) based on a positive anti-cyclic citrullinated peptide 2 (CCP2) antibody test, and having musculoskeletal complaints but no signs of arthritis. Study participants were followed for ≥3 years (arthritis-free, n=165) or until arthritis onset (progressors, n=95). Anti-Rgp IgG was measured in Risk-RA (260 baseline and 247 follow-up samples) and healthy controls (n=126); anti-CCP2 IgG was measured in Risk-RA (254 baseline samples). Data were analysed in GraphPad Prism and R using log-transformed antibody levels.

Results: 53% of Risk-RA and 26% of controls, p=0.003, were anti-Rgp IgG positive at baseline, with higher levels in Risk-RA compared with controls, p<0.0001. No changes in anti-Rgp IgG levels were observed during follow-up. The anti-Rgp IgG response at baseline did not associate with the development of arthritis; Cox-regression showed an HR of 0.95 (CI 0.80 to 1.13, p=0.6) for anti-Rgp IgG levels, and 0.82 (CI 0.55 to 1.23, p=0.3) for anti-Rgp IgG positivity.

Conclusions: Antibodies against the oral bacterium are elevated during the RA-risk phase, both in individuals progressing to arthritis and in individuals remaining arthritis-free. Hence, infection can be linked to the presence of RA-specific autoimmunity, ACPA, and musculoskeletal symptoms, but not to further development of arthritis in this at-risk population.