Development and optimization of sensitive peptide elisa for autoantibody testing in early and seronegative rheumatoid arthritis

Abstract

Background Using a cDNA phage display approach, 14 novel autoantibody markers were previously discovered that can be detected in early and seronegative (rheumatoid factor-negative, anti-cyclic-citrullinated-peptides-negative) RA-patients1. The level of autoantibody reactivity of individual plasma samples was measured using phage enzyme-linked immunosorbent assays (ELISAs) demonstrating a combined sensitivity of 54% and an associated specificity of 90% for RA. For UH-RA.21, the autoantibody marker constituting the highest sensitivity (29%), a sensitive peptide ELISA had already been developed, easy reproducible and fit for large scale screening2.

Objectives Our goal is to convert the phage ELISAs used for initial screening into sensitive peptide ELISAs for the remaining markers of the autoantibody panel as well. Furthermore, we aim at validating the candidate markers through testing of autoantibody reactivity in a patient cohort.

Methods Following optimization of a solid phase ELISA format, antibody reactivity was measured in a pilot cohort consisting of up to 57 RA-patients, 74 healthy subjects (HC) and 30 patients suffering other rheumatic diseases (RC). Further validation of the candidate markers is currently being performed through screening of established RA-patients at 3 different time points (n=192) together with additional HC (n>40). Moreover, another cohort is at our disposal consisting of undifferentiated arthritis patients, early RA-patients (symptoms <1 year) and non-healthy controls including patients with other inflammatory (non)-rheumatic diseases and individuals with mechanical joint complaints.

Results Peptide ELISAs were developed and optimized for autoantibody testing for UH-RA.1, UH-RA.9 and UH-RA.14. Screening of antibody reactivity against UH-RA.1 and UH-RA.9 in a pilot cohort demonstrated a sensitivity of 6,3% (3/48) for UH-RA.9 with an associated specificity of 100% as 0/50 HC and 0/24 RC tested positive on this candidate marker; for UH-RA.1 a sensitivity of 8,8% (5/57) was shown with a specificity of 93% as 4/74 HC and 3/30 RC also showed an increased antibody reactivity. No individual plasma samples tested positive on both UH-RA.1 and UH-RA.9. The sensitivities and specificities for both markers were comparable to those from the initial cohort screened by phage ELISA. Using both phage and peptide ELISAs, increased antibody reactivity could be detected in seronegative patients. For UH-RA.14 screening of the pilot cohort has started up very recently, and the development of peptide ELISAs for the remaining markers of the panel is still ongoing. Screening results from the pilot cohort are currently validated in an additional cohort in which the potential of the novel biomarkers in diagnosing early and seronegative RA-patients is further investigated. Furthermore, the prognostic value is evaluated by studying possible correlations with clinical outcome, disease parameters and response to therapy.

Conclusions Peptide ELISAs were successfully developed and optimized for UH-RA.1, UH-RA.9 and UH-RA.14. The application of these peptide ELISAs in screening of additional plasma samples allows further validation of diagnostic and prognostic potential of our novel biomarkers.