A deep dive into four thyroglobulin immunoassays from analytical perspective

Serum thyroglobulin immunometric assays (sTg) are crucial for monitoring differentiated thyroid cancer (DTC) treatment. However, challenges such as anti-thyroglobulin autoantibodies (TgAb) and assay variability hinder evaluations. This study assessed four sTg methods—three second-generation (Architect, Access, Elecsys) and one first-generation (Immulite)—following Clinical and Laboratory Standards Institute (CLSI) and American Thyroid Association (ATA) guidelines.
The study compared sTg(Architect), sTg(Access), sTg(Elecsys), and sTg(Immulite). Precision was evaluated per CLSI EP05-A3, while the lower limits of detection (LLD) were assessed using EP17-A2. Passing–Bablok and Bland–Altman analyses were conducted as per EP09c, and semi-quantitative comparisons used Kappa statistics.
The second-generation sTgs (Architect, Access, Elecsys) exhibited satisfactory precision (<7% coefficient of variation, CV%), unlike sTg(Immulite), which showed significant deviations and inadequate sensitivity for DTC recurrence (Limit of quantitation, LoQ = 4.59 μg/L). Second-generation sTgs had strong correlations (r > 0.884) across all concentration ranges (≤1, 1-10, >10 μg/L), with biases (slope: 1.131-2.027). sTg(Immulite) correlated well with second-generation methods for concentrations >10 μg/L (r > 0.945) but less so for <10 μg/L (r < 0.642). TgAb significantly impacted sTg(Immulite). Kappa statistics revealed strong agreement among second-generation methods (κ > 0.800) but lower concordance with sTg(Immulite), especially in TgAb(+) samples (κ: 0.562-0.653). Agreement ratios were high for second-generation methods (0.667-1.000) but variable for sTg(Immulite), particularly at lower concentrations and in TgAb(+) cases (0.097-0.727).
sTg(Immulite) did not meet LLD and precision criteria for DTC monitoring, facing issues with TgAb interference. Second-generation sTgs demonstrated consistent performance across all concentrations.

血清甲状腺球蛋白免疫测定法（serum thyroglobulin immunometric assays，sTg）是监测分化型甲状腺癌（differentiated thyroid cancer，DTC）治疗转归的关键手段。然而，抗甲状腺球蛋白自身抗体（anti-thyroglobulin autoantibodies，TgAb）及检测方法变异性等问题，均对临床评估构成了阻碍。
本研究遵循临床和实验室标准协会（Clinical and Laboratory Standards Institute，CLSI）与美国甲状腺协会（American Thyroid Association，ATA）发布的指南，对4种sTg检测方法开展评估——其中3种为第二代检测平台（Architect、Access、Elecsys），另1种为第一代检测平台（Immulite）。
本研究对sTg(Architect)、sTg(Access)、sTg(Elecsys)及sTg(Immulite)四种检测方法进行了头对头比较。按照CLSI EP05-A3标准评估检测精密度，采用EP17-A2标准测定最低检测限（lower limits of detection，LLD），依据EP09c标准实施Passing-Bablok与Bland-Altman分析，并通过Kappa统计量完成半定量比较。
第二代sTg检测平台（Architect、Access、Elecsys）展现出良好的精密度（变异系数<7%，CV%），而sTg(Immulite)则存在显著偏差，且对DTC复发的检测灵敏度欠佳（定量限Limit of quantitation，LoQ=4.59 μg/L）。第二代sTg检测平台在所有浓度区间（≤1、1~10、>10 μg/L）均表现出较强相关性（r>0.884），偏倚斜率介于1.131~2.027之间。sTg(Immulite)在浓度>10 μg/L时与第二代检测平台相关性良好（r>0.945），但在浓度<10 μg/L时相关性较弱（r<0.642）。TgAb会对sTg(Immulite)产生显著干扰。Kappa统计结果显示，第二代检测平台之间一致性较强（κ>0.800），但与sTg(Immulite)的一致性较低，尤其在TgAb阳性（TgAb(+)）样本中（κ：0.562~0.653）。第二代检测平台的一致性比率较高（0.667~1.000），但sTg(Immulite)的一致率波动显著，尤其是在低浓度样本及TgAb(+)样本中（0.097~0.727）。
sTg(Immulite)未满足DTC监测所需的最低检测限与精密度标准，且存在TgAb干扰问题。第二代sTg检测平台在所有浓度区间均展现出稳定可靠的检测性能。