Background: We validated the DiaSorin Liaison Calcitonin_II-Gen, an improved method for calcitonin (CT) measurements, compared this method with the Cisbio_h-CT kit and established the reference range of CT in a normal adult population.
Methods: We determined the precision, functional sensitivity, traceability to the 2nd IS 89/620, linearity and measurement uncertainty, accuracy profile and β-expectation limits. We evaluated the specificity, the susceptibility to human anti-animal antibodies (HAMA), hook-effect and carry over. To establish a reference range, we selected 267 adults without renal insufficiency presenting with normal thyroid stimulating hormone (TSH), free thyroxin (T4) and calcium concentrations and without anti-thyroglobulin antibodies as our "reference" healthy population. We compared the method with Cisbio on 250 consecutive and 45 samples from a post-pentagastrin stimulation test.
Results: Precision (expressed as CV) was < 10% for the measurement range, functional sensitivity: 5.3 ng/L and the method was found linear until to a 1/10 dilution. Uncertainty ranged from 25% to 7.2%, and the risk that one result falls out of the ± 20% acceptance limits was < 5% between 2.9 and 1513 ng/L. The Bland and Altman plot showed no systematic bias between the two methods. The test is still prone to HAMA influence, does not present any hook-effect, although carry over was observed. Ninety-five percent of our adult reference population showed CT concentrations < 7.4 ng/L, with an important gender difference: 95% of the men showed CT values < 9.8 ng/L, whereas 95% of women were < 4.0 ng/L.
Conclusions: The Liaison Calcitonin_II-Gen is an analytically robust method. The important difference in gender observed in our population might lead to re-evaluation of the generally used "10 ng/L" cut-off in a multicentre prospective study.