Background/Purpose:

Anti-citrullinated protein antibodies (ACPA) are predictive markers with pathological effects in rheumatoid arthritis (RA) development. Previous studies of ACPA-positive patients with musculoskeletal complaints (mainly arthralgia), but no arthritis, have shown increased risk for arthritis onset. However, these studies used clinical definition of arthritis, but did not exclude those subjects with ultrasound-detected subclinical arthritis. We aimed therefore to investigate risk factors for developing arthritis in ACPA-positive subjects with musculoskeletal complaints in the absence of any clinical and ultrasound signs of arthritis.

Methods:

Patients presenting with musculoskeletal complaints and positive Anti-citrullinated protein antibody (ACPA) test in primary care were referred to the Rheumatology Clinic at Karolinska Hospital. Patients lacking arthritis by clinical and ultrasound examination (defined as synovial hypertrophy with Doppler activity) were recruited into the Risk-RA research program. A total of 66 subjects included between years 2015 up to December 2016 were analysed in this study. Blood samples from inclusion have been analysed for thirteen ACPA reactivities (citrullinated peptides from fillagrin, fibrinogen, alfa-enolase, vimentin, histones) using microarray based on ImmunoCap ISAC. DNA samples have been analysed for HLA-SE risk gene using DR low-resolution kit (2-digit).

Results:

A high proportion of the Risk-RA subjects, 41% (27 out of 66), developed clinical and/or ultrasound-detected arthritis during a median follow up of 8 months while 59% (39 out of 66) subjects didn’t developed arthritis during a median follow up of 25 months.

No differences in characteristics at inclusion were observed between subjects developing and those not developing clinical and/or ultrasound-detected arthritis (table 1). Interestingly, subjects developing arthritis had higher concentration of anti-CCP and number of tender joints as compared to those not developing arthritis, but the difference did not reach statistically significance.  

In contrast, subjects developing arthritis had a higher number of ACPA-reactivities (mean 6) than those not developing arthritis (mean 3, t test, p< 0.05). A significant difference in occurrence of a HLA-SE risk-gene was detected with subjects developing arthritis being more often carrier of such risk-gene (86%) as compared to those not developing arthritis (56%, chi-square test, p<0.05). Cox proportional hazards regression accounting for the time of follow-up showed a HR for arthritis development of 1.1 for every increase in number of ACPA reactivities (95% CI 0.99-1.2, p 0.07) and 4.9 (95% CI 1.5-16, p 0.01) for HLA-SE carriers.

Conclusion:

Subjects with ACPA-positive musculoskeletal complaints lacking any clinical and ultrasound signs of arthritis are at high risk to develop arthritis, especially among carriers of HLA-SE risk-gene.

Table 1.