Abstract
Background/Objectives: Anti-cyclic citrullinated peptide (anti-CCP) antibodies are highly specific markers for rheumatoid arthritis (RA). Over the past decade, novel automating detection systems have been developed for anti-CCP detection. The present study aimed to evaluate the diagnostic performances of three fully automated anti-CCP assays in comparison to a conventional manual enzyme-linked immunosorbent assay (ELISA). Methods: One hundred ninety-nine patients with rheumatic symptoms (100 with RA and 99 without RA) were tested for anti-CCP autoantibodies using four assays: a manual-ELISA (EUROIMMUN(®)), two chemiluminescence immunoassays (CLIAs) performed on the MAGLUMI X3(®) and iFlash 1800(®) platforms, and an enzyme immunoassay (EIA) run on the UNI(®) analyzer. Results: The Kappa statistic indicated a moderate qualitative agreement among the EUROIMMUN, iFlash, and UNI assays (0.734 ≤ ĸ ≤ 0.778), while the MAGLUMI anti-CCP assay showed only weak-to-moderate agreement with the others (0.510 ≤ ĸ ≤ 0.628). A strong positive correlation was observed between anti-CCP levels measured by the four assays (0.747 ≤ rho ≤ 0.839). At the manufacturers' cut-off values, sensitivities ranged from 76% to 99% and specificities from 69.7% to 99%, depending on the assay. However, at a fixed specificity of 95%, all the four assays showed good diagnostic performances for RA, with sensitivities ranging from 80% to 89% and positive likelihood ratios (LRs+) from 16 to 17.8. Conclusions: Our results revealed that at the manufacturers' cut-offs, the UNI anti-CCP assay was the most valuable alternative to the conventional ELISA for diagnosing RA in our cohort. Nevertheless, after an appropriate adjustment of the thresholds, all the evaluated assays showed good diagnostic performances for RA.