Title: Categorization of Rheumatoid Arthritis Subjects by Rheumatoid Factor and Anti-Cyclic Citrullinated Autoantibody Status Identifies Rheumatoid Arthritis Subjects with Different Clinical Characteristics
Swati Modi1, Mariely Nieves-Plaza2, Donald M. Jones1, Erich R. Wilkerson1, Christine L. Amity1, Kelly A. Reckley1, Ilinca Metes1, Jason Lyons3, Heather Eng3, Stephen R. Wisniewski3 and Marc C. Levesque1
University of Pittsburgh, Pittsburgh, PA; 1 School of Medicine, Department of Medicine, Division of Rheumatology and Clinical Immunology, 2 Clinical and Translational Science Institute, 3 School of Public Health, Department of Epidemiology
Background/Purpose: Rheumatoid factor (RF) and anti-citrullinated cyclic peptide (CCP) antibodies have been used to diagnose rheumatoid arthritis (RA) patients, although there is substantial heterogeneity among RA patients regarding the presence of these biomarkers. Previously, we found that RA patients grouped on the basis of RF and CCP status into 4 groups: RF+CCP+, RF+CCP-, RF-CCP+, RF-CCP- had different clinical characteristics including a higher percentage of female subjects and less use of biologics in the RF+CCP- group. Our aim was to determine whether these subgroups of RA patients had features suggestive of other overlapping connective tissue diseases.
Methods: Patients from the Rheumatoid Arthritis Comparative Effectiveness Research (RACER) registry who met the 1987 and/or 2010 ACR criteria for a diagnosis of RA were categorized based on clinical cut-offs for RF and CCP into 4 groups: RF+CCP+, RF+CCP-, RF-CCP+, RF-CCP-. We determined the association of these four RF and CCP groups with autoantibodies and inflammatory markers measured by ELISA (IL-6) or in a clinical laboratory (ANA, anti-SSA/SSB, anti-dsDNA, C3, C4, gammaglobulin and CRP). Categorical and continuous variables were analyzed using chi-square and Kruskal-Wallis tests, respectively.
Results: IL-6 levels were significantly associated with disease activity (DAS28-CRP) but only in the RF+/CCP+ and RF-/CCP- groups, while CRP levels were significantly associated with disease activity in all groups. ANA positivity was significantly different across groups (p = 0.008) (Table 1). The percentage of ANA positive subjects was highest in the RF+CCP+ group (71%) and lowest among the RF-/CCP- group (48%). There were no significant differences for the remaining laboratory characteristics, although RF+/CCP- subjects were more likely to have a low C4 level and a positive anti-SSA and/or -SSB and anti-dsDNA.
Conclusion: RF+CCP- subjects were typically ANA positive and more likely than other RF/CCP groups to be female, have low C4 levels and to have a positive anti-SSA and/or -SSB suggesting that some of these subjects may have RA and/or Sjogren’s syndrome. The lower percentage of subjects in the CCP negative groups treated with biologic therapies may be due to lower levels of disease activity as seen in the RF-CCP- group and/or may be due to the presence of subjects with diseases poorly responsive to biologic therapies such as Sjogren’s syndrome in the RF+/CCP- group. In summary, we believe that categorization of RA subjects by RF and CCP status may identify RA subjects with different syndromes, differential responses to therapy and different outcomes.