B-104 Comparison of the Performance Characteristics of the Third Generation Anti-Cyclic Citrullinated Peptide CCP3.1 IgG/IgA and CCP3 IgG Assays

Abstract

Abstract Background Rheumatoid arthritis (RA) is one of the most common systemic autoimmune diseases. Anti-citrullinated protein antibody (ACPA) is the most specific serological biomarker for RA and is associated with significantly more disease activity compared to rheumatoid factor (RF). Improved autoantibody detection relied in the generation of tests against cyclic citrullinated peptide (CCP), a subset of ACPA. The 3rd generation anti-CCP IgG antibody (CCP3) emerged with increased sensitivity compared to the earlier versions. More recently, CCP3.1 with conjugate to detect IgA antibodies in addition to the usual IgG antibodies became available. This is described to exhibit enhanced sensitivity due to early RA patients producing IgA antibodies to CCP3 in the absence of IgG. In this study, we evaluated the diagnostic characteristics of CCP3.1 and its association with disease severity in comparison to CCP3. Methods Our cohort included a total of 331 subjects submitted for RA panel or independent antibody testing. Chart review revealed 136 clinically defined RA- positive and 195 RA- negative patients. Residual sera from these patients were tested for RF (Roche Diagnostics, Indianapolis, US), CCP3 and CCP3.1 (Inova Diagnostics, San Diego, US) antibodies. Clinical performance characteristics at the manufacture-suggested cut-off and association with disease severity were assessed and compared between the two assays. Results The overall diagnostic accuracy was comparable between the two assays with CCP3 exhibiting an area under the curve (AUC) of 0.879, 95% CI: 0.836–0.922, and CCP3.1 showed an AUC of 0.890, 95% CI: 0.849–0.931. CCP3 displayed a sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of 78.7%, 90.8%, 85.6% and 85.9% respectively. CCP 3.1 displayed an equivalent performance with 77.9% sensitivity, 93.3% specificity, 89.1% PPV and 85.8% NPV. Both assays showed an excellent agreement with a positive percent agreement of 92.0% and negative percent agreement of 98.7%. There was a strong quantitative correlation between the two assays (Spearman correlation coefficient, r = 0.864, P < 0.001). Paired t test showed significantly higher titers of CCP3 (mean:118, median:145) compared to CCP3.1 (mean:110, median: 108) in RA patients. RF and CCP3.1 combination yielded a slightly greater specificity (99.0%), positive likelihood ratio (65.9) and odds ratio (201.7) compared to the RF and CCP3 combination (specificity: 98.5%, positive likelihood ratio: 44.5, odds ratio: 138.4). Nine early-RA patients (with symptom onset <6 months of diagnosis) were detected by both assays and displayed comparable titers (CCP3.1, mean:138, median:149; CCP3, mean:149, median:161). Out of 60 RA patients exhibiting more severe disease course with the presence of bone erosions, 50 of them were detected by CCP3.1 (mean:125, median:146) and 51 were detected by CCP3 (mean:139, median:151) and the titers were comparable. Conclusion CCP3 and CCP3.1 assays displayed comparable performance for detection of RA and correlated similarly with the disease severity. In our study cohort that mainly comprised of RA patients diagnosed >6 months after symptoms onset, addition of IgA isotype showed no added value but did not compromise performance.