Search for a command to run...
Abstract Background Despite significant improvements in the treatment and management of Axial Spondyloarthritis (AxSpA), diagnosis is often delayed by 5-10 years. Current diagnostic practices rely on imaging like MRI and HLA-B27 typing, which are not readily available to most referring physicians. Autoantibodies to 14-3-3eta antigen represent a promising biomarker. This study aims to evaluate the technical performance and robustness of the 14-3-3eta AAb multiplex assay, demonstrate its clinical performance in AxSpA compared to healthy subjects, and generate a multi-analyte algorithm to increase diagnostic accuracy. Methods To evaluate the 14-3-3eta AAb assay, we assessed precision, LoB, LoD, sensitivity, specificity, linearity, hook effect, interference, accuracy, and robustness. ROC curve analysis and Youden index established peptide positivity cut-offs. Standards quantified AAb levels in serum. A composite score based on peptide positivity was developed to compare axSpA diagnosis likelihood (n = 83) to healthy subjects (n = 57). Results The 14-3-3eta AAb assay demonstrated high precision, with intra-assay %CVs ranging from 4.5% to 11.4% and intra-laboratory precision %CVs ranging from 8.3% to 14%. The LoD was on average 3.2X the LoB, with the positivity cut-off at 2.3X the LoD, and 6.5X the LoB. The hook effect was evaluated at concentrations significantly higher than physiologically probable, and all targets did not show a meaningful loss of signal at high concentrations. Accuracy was high, with a percentage agreement of 96.0%. All peptides demonstrated positive predictive values (PPVs) ranging from 70.6% to 83.3%, exceeding the acceptance criteria. The assay showed minimal cross-reactivity and interference from normal serum elements (hemoglobin, bilirubin, triglycerides, cholesterol, and albumin), non-target antibodies (HAMA, heterophile, and infliximab), and common therapies (ibuprofen, sulfasalazine, and dexamethasone). All assay targets demonstrated linear dose-dependent signals (1.00 ± 0.05). The 14-3-3eta AAb assay yielded strong performance despite stress testing through protocol deviations, such as delays in secondary antibody incubation and final reading, confirming its robustness. As presented in Table 1, prioritized peptides 1 - 5 yielded significant areas under the curve (AUC) when discriminating axSpA from presumed healthy controls. Using cut-offs derived from the Youden index for each peptide, positivity scores were assigned and utilized to generate a composite positivity score. The strength of the association for an axSpA diagnosis was evaluated using the Fisher’s Exact test, with the model yielding a Chi-Square of 31.8, p < 0.0001 with an Odds Ratio of 8.7 (95% CI 3.9 - 19.6). The relative risk for each group (axSpA or healthy) based on composite positivity status is presented in Table 2. The data demonstrates a patient’s higher relative risk for axSpA based on being composite positive when compared to healthy subjects. Conclusion The 14-3-3eta AAb multiplex assay exhibits excellent technical performance, characterized by high precision, accuracy, and robustness. It successfully differentiates between axSpA patients and healthy controls, showing minimal cross-reactivity and interference. The composite score based on peptide positivity significantly improves diagnostic confidence for axSpA. These results underscore the assay*s reliability and clinical utility. Future research will investigate AAb expression in various autoimmune conditions.