United States and European Multicenter Prospective Study for the Analytical Performance and Clinical Validation of a Novel Sensitive Fully Automated Immunoassay for Calcitonin

Abstract

Background: The objective of this study is the validation and proof of clinical relevance of a novel electrochemiluminescence immunoassay (ECLIA) for the determination of serum calcitonin (CT) in patients with medullary thyroid carcinoma (MTC) and in different diseases of the thyroid and of calcium homeostasis.

Methods: This was a multicenter prospective study on basal serum CT concentrations performed in 9 US and European referral institutions. In addition, stimulated CT concentrations were measured in 50 healthy volunteers after intravenous calcium administration (2.5 mg/kg bodyweight).

Results: In total, 1929 patients and healthy controls were included. Limits of blank, detection, and quantification for the ECLIA were 0.3, 0.5, and 1 ng/L, respectively. Highest intra- and interassay coefficients of variation were 7.4% (CT concentration, 0.8 ng/L) and 7.0% (1.1 ng/L), respectively. Medians (interval) of serum CT concentrations in 783 healthy controls were 0.8 ng/L (<0.5-12.7) and 3 ng/L (<0.5-18) for females and males, respectively (97.5th percentile, 6.8 and 11.6 ng/L, respectively). Diagnostic sensitivity and specificity were 100%/97.1% and 96.2%/96.4%, for female/males, respectively. Patients (male/female) with primary hyperparathyroidism, renal failure, and neuroendocrine tumors showed CT concentrations >97.5th percentile in 33%/4.7%, 18.5%/10%, and 8.3%/12%, females/males, respectively. Peak serum CT concentrations were reached 2 min after calcium administration (161.7 and 111.8 ng/L in males and females, respectively; P < 0.001).

Conclusions: Excellent analytical performance, low interindividual variability, and low impact of confounders for increased CT concentrations in non-MTC patients indicate that the investigated assay has appropriate clinical utility. Calcium-stimulated CT results suggest good test applicability owing to low interindividual variability.