Background/Purpose: Rheumatoid arthritis (RA)-related autoantibody (Ab) testing for Abs to cyclic citrullinated peptide (CCP) and rheumatoid factor (RF) are used in the clinical diagnosis of RA. In addition, RA-related Ab elevations in asymptomatic individuals indicate an increased risk of future RA and define a preclinical period of RA development. These at-risk subjects could be considered for prevention strategies in RA, but a challenge in effective identification of individuals at-risk for RA is the varied methods of Ab testing. For example, RF can be tested by nephelometry (RFneph) as is commonly used in clinical labs or by ELISA for RF isotypes (IgM/A/G). Studies demonstrate RFIgM and/or RFIgA is highly sensitive in classifiable RA, but the utility of RF isotype testing in addition to RFneph is not well known in preclinical RA.

Methods: We evaluated subjects with classifiable RA (1987 ACR criteria) from the Studies of the Etiology of RA (SERA) project (N=566) and the Department of Defense Serum Repository (DoDSR; N=83; all with pre-diagnosis and 47/83 with post-diagnosis samples). Controls were 200 random blood donors for SERA RA cases and 83 DoDSR matched controls for DoDSR RA cases. All subjects were tested for RFneph and RF isotypes (IgM/A/G by ELISA). Cut-off levels for all RF assays were determined as <5% positive in a separate set of 491 blood donors. CCP3.1 (IgG/IgA INOVA) testing was performed in all subjects using manufacturers cut-off levels.

Results: The diagnostic accuracy of Ab testing is listed in Table 1. All RFs and CCP3.1 had high specificity for classifiable RA including SERA RA and DoDSR post-diagnosis. In DoDSR pre-diagnosis testing, sensitivity for future RA was 41% for RFneph and 63% for CCP3.1. Sensitivity increased to 69% when positivity for the following Abs was considered: RFneph, CCP3.1, RFIgA or RFIgM. In DoDSR RA cases who were negative for RFneph and CCP3.1 in pre-diagnosis testing (N=29), adding RFIgA or RFIgM (but not IgG) testing identified an additional 10% of subjects who developed classifiable RA (Table 2).

Conclusion: Herein RFneph and CCP3.1 had fair diagnostic accuracy for future RA; however, RFIgA and IgM (but not IgG) testing allowed for identification of 10% more individuals who would develop future RA, while maintaining high specificity (93% for RFIgA or IgM pre-diagnosis). These findings suggest that RFneph misses some RF positivity that is detectable by isotype testing. As such, in addition to CCP3.1 and RFneph, RFIgM and RFIgA testing may be clinically meaningful in identifying subjects at-risk for future RA, especially if interventions become available to prevent progression from preclinical to classifiable RA.