The association between anti-carbamylated protein (anti-CarP) antibodies and radiographic progression in early rheumatoid arthritis: a study exploring replication and the added value to ACPA and rheumatoid factor

Objective Anti-carbamylated protein (anti-CarP) antibodies are reported to associate with more radiographic progression within the total rheumatoid arthritis (RA) population and anti-citrullinated peptide antibody (ACPA)-negative subgroup. We explored the association of anti-CarP with radiographic progression in RA and aimed to replicate the association and evaluate the added value of anti-CarP antibodies in relation to ACPA and rheumatoid factor (RF). Methods 576 Swedish and 628 Dutch patients with RA (2394 and 3247 sets of radiographs, respectively) were longitudinally studied. Replication was restricted to the Swedish patients. In both cohorts, the association of anti-CarP with radiographic progression was determined in strata of patients with similar ACPA and RF status; results of both cohorts were combined in fixed-effect meta-analyses. The net percentage of patients for whom the radiographic progression in 5 years was additionally correctly classified when adding anti-CarP to a model including ACPA and RF was evaluated. Results Anti-CarP associated with radiographic progression in the total Swedish RA population (beta=1.11 per year, p=8.75x10(-13)) and in the ACPA-negative subgroup (beta=1.14 per year, p=0.034). Anti-CarP associated with more radiographic progression in the strata of ACPA-positive/RF-negative, ACPA-negative/RF-positive and ACPA-positive/RF-positive patients with RA (respective p values 0.014, 0.019 and 0.0056). A model including ACPA and RF correctly classified 54% and 57% of the patients; adding anti-CarP to this model did not increase these percentages (54% and 56% were correctly classified). Conclusions Anti-CarP antibodies associated with more severe radiographic progression in the total and ACPA-negative RA population. Anti-CarP -positivity had a statistically significant additive value to ACPA and RF, but did not improve correct classification of patients.