Background/Purpose: Anti-cyclic citrullinated peptide-2 (CCP2) can be present in serum years before a diagnosis of rheumatoid arthritis (RA), and is highly specific (>95%) for the future onset of RA: however, its sensitivity for future RA is relatively low. Thus, novel autoantibodies (Abs) that are highly sensitive and specific for RA would be a valuable predictive tool. Recently, Shi and colleagues (Shi et al, PNAS, 2011) discovered that anti-carbamylated protein (anti-CarP) was not sensitive (43%), but highly specific (95%) for classified RA as well as present prior to the onset of classifiable RA in subjects with arthralgia but no inflammatory arthritis who were positive for CCP and/or rheumatoid factor (RF)(IgM) (Shi et al, A&R, 2013). Herein we utilized a unique sample set of military subjects to determine if serum anti-CarP antibodies are present before RA diagnosis, and compared the diagnostic accuracy of anti-CarP to other RA-related Abs.

Methods: Stored pre-diagnosis serum samples from military personnel were obtained from 82 RA cases and 82 controls matched on case age at diagnosis, sex, and race. We tested the pre-RA diagnosis samples from cases and controls for anti-CarP Fetal Calf Serum (FCS) and anti-CarP Fibrinogen (Fib), as well as RF by nephelometry, RF isotypes (IgM, IgG, and IgA), CCP2 (Axis-Shield), and CCP3.1 (Inova). The cutoffs for anti-CarP were determined using the mean +2 SD of log-transformed levels from 41 randomly selected military controls. The diagnostic accuracy of Abs for future RA were assessed in prediagnosis serum and logistic regression was used to determine the association between RA case status and antibody positivity in prediagnosis serum, comparing RA cases to the 41 remaining controls.

Results: RA cases, cutoff controls, and controls had similar demographic characteristics. The diagnostic accuracy in prediagnosis serum for each Ab is presented in the Table. Anti-CarP FCS was 29.9% sensitive (SENS) and 95% specific (SPEC) for future RA; anti-CarP Fib was 17.9% SENS and 95.1% SPEC. There was a strong association between RF and/or CCP positive RA case status and anti-CarP FCS positivity in prediagnosis serum (OR: 8.3, 95% CI: 1.8-37.7, p<0.01); however, there was a non-significant association between seropositive RA case status and anti-CarP Fib positivity in prediagnosis serum (OR: 4.3, 95% CI: 0.9-20.8, p=0.07).

Conclusion: Our results suggest that anti-CarP antibodies, are present in RA cases before clinically apparent disease, and the presence of anti-CarP FCS is strongly associated with future onset of RA (Positive Predictive Value=90.9%). Identification of novel antibodies such as anti-CarP FCS can elucidate disease mechanisms and predict the onset of RA.