Comparative evaluation of two automated immunoassays for serum thyroglobulin quantification

Background: We evaluated the clinical performance of in vitro diagnostic devices for quantifying thyroglobulin (Tg), which is a key marker for monitoring and treating thyroid dysfunction. The recently launched Siemens Atellica IM Tg assay was compared with the established Roche Elecsys Tg II assay using residual serum samples from routine testing.Methods: The precision, linearity, limit of blank (LoB), limit of detection (LoD), limit of quantitation (LoQ), and reference ranges were assessed using the Siemens Atellica IM Tg assay. In total, 681 patient serum samples were analyzed to compare the results with those of the Roche Elecsys Tg II assay for correlation and concordance evaluations across the clinical ranges.Results: The precision coefficients of variation, linearity, LoB, LoD, and LoQ for the Siemens Atellica IM Tg assay met the manufacturer’s specifications across all concentrations. All data points and confidence intervals were within the allowable deviation from linearity. The correlation between the assays was excellent (Pearson’s r=0.997). In the low range, concordance was 83%, whereas in the normal range, it reached 98%. In contrast, the highly abnormal range exhibited a concordance of 65%, resulting in an overall concordance rate of 88%. Weighted kappa values (κ=0.79–0.82) demonstrated moderate-to-strong agreement.Conclusions: The Siemens Atellica IM Tg assay showed performance consistent with the precision, linearity, LoB, LoD, LoQ, and reference ranges claimed by the manufacturer. It demonstrated a strong correlation and good overall concordance with Roche Elecsys Tg II. Lower concordance in the highly abnormal range suggests a potential limitation.