Diagnostic performance of the ARCHITECT C-Peptide immunoassay

Background: Measurement of C-peptide under standardized conditions provides a sensitive and well-established assessment of β-cell function. We describe the analytical and clinical validation of an automated, microparticle-based chemiluminescent immunoassay method. The assay is designed to measure C-peptide in human serum, plasma and urine. Methods: Assay performance characteristics such as precision and recovery were measured according to protocols established by the Clinical Laboratory Standards Institute (CLSI). A reference range study was conducted. Analytical sensitivity and specificity, interfering substances, recovery, and linearity studies were performed. Method comparison, against the ADVIA Centaur C-Peptide assay (Siemens), was evaluated with clinical specimens from patients with abnormal insulin secretion. Results: The detection limit for this assay was 0.01 ng/mL. Functional sensitivity (inter-assay imprecision ≤20%) was 0.015 ng/mL at a coefficient of variation (%CV) of 11.2%. Total imprecision was below 6.5% CV. The assay was linear upon dilution. Comparison with the ADVIA Centaur C-Peptide assay yielded a correlation coefficient (r) of 0.99. Conclusions: The ARCHITECT C-Peptide assay measures C-peptide rapidly, accurately, and precisely in human serum, plasma and urine. It provides useful improvements for β-cell function testing and for evaluating the clinical status of a patient in combination with other diabetes markers. Clin Chem Lab Med 2009;47:834–41.Peer Reviewe