Background/Purpose: Previous longitudinal studies in arthralgia patients studied Rheumatoid Factor (IgM-RF) and Anti-Citrullinated Protein Antibodies (ACPA) in relation to arthritis development. Participants in these studies were often selected on the presence of at least one autoantibody and consequently the predictive values of different autoantibodies were evaluated relative to each other. This study evaluated the association of RF, ACPA and anti-carbamylated protein antibodies and arthritis development in arthralgia patients that were selected based on the clinical suspicion to progress to RA (Clinically Suspect Arthralgia).

Methods: 255 Clinically Suspect Arthralgia patients were included in our cohort. Patients had arthralgia for <1 year, without clinical arthritis and were considered at risk for RA based on clinical presentation by their rheumatologists. RF, ACPA and anti-carbamylated protein status determined at baseline were studied in relation to the development of clinical arthritis during a median follow-up of 96 week using univariable and multivariable Cox regression analyses.

Results: 45 patients developed clinical arthritis. In univariable Cox regression analyses, presence of RF (HR=4.8, 95%CI=2.7-8.7), ACPA (HR=7.9, 95%CI=4.4-14.3) and anti-carbamylated protein (HR=3.7, 95%CI=1.9-7.3) antibodies were all shown to have a significant association with arthritis development. In multivariable Cox regression including RF, ACPA and anti-carbamylated protein, only ACPA-seropositivity was significantly associated with arthritis development (HR=5.0, 95%CI=1.9-12.9). When stratifying the patients in groups with different combinations of RF and ACPA (see Figure 1), Cox regression revealed the highest hazard ratio for RF-positive ACPA-positive patients (HR=9.5, 95%CI=4.9-18.3, compared to RF-negative ACPA-negative patients). No significant differences were observed for RF-negative ACPA-positive compared to RF-positive ACPA-positive patients. Although nearly 70% of ACPA-positive RF-positive patients progressed to clinical arthritis, 30% did not develop clinical arthritis during a median follow-up duration of 96 weeks.

Conclusion: Within Clinically Suspect Arthralgia patients, presence of ACPA alone or in combination with RF, is associated with an increased hazard on progression to clinical arthritis. However, although still 30% of patients that have both antibodies did not develop arthritis within 96 weeks follow-up and patients without antibodies also developed arthritis suggest that information on autoantibodies alone is not sufficient for optimal risk stratification.

Figure 1. Kaplan-Meier One Minus Survival plots for combinations of Rheumatoid Factor and Anti-Citrullinated Protein antibodies and progression to clinical arthritis. Glossary: Kaplan-Meier one minus survival plot showing cumulative progression to clinical arthritis for Clinically Suspect Arthralgia patients, according to being positive for either positive for anti-citrullinated protein antibodies (ACPA) and rheumatoid factor (RF) positive (N=28), single ACPA-positive (N=6), single RF-positive (N=26) and double negative patients (N=195). Cox regression of combinations of RF- and ACPA-positivity revealed that the presence of both ACPA and RF yielded the highest HR of 9.5 (95%CI=4.9–18.3) for arthritis development compared to the group negative for both autoantibodies, as shown in figure 1. The hazard ratio of single ACPA-positivity was 8.1 (95%CI=2.4–27.4), compared to the group that was negative for both RF and ACPA. The hazard ratio for single RF-positivity was 2.5 (95%CI=1.0–6.3) compared to the group that was negative for both RF and ACPA.