B-187 Evaluation of Clinical Utility of LC-MS/MS Method for Thyroglobulin Measurement in Comparison with Immunoradiometric Assay and Chemiluminescence Microparticle Immunoassay

Abstract Background Thyroglobulin (Tg) is the follow-up index of choice for patients with differentiated thyroid cancer (DTC), especially who were undergone total thyroidectomy. However, due to the commonly used immunoassay’s inherent susceptibility to interference combined with the high prevalence of thyroglobulin antibody (TgAb) in DTC patients, the need for more reliable measurement for Tg independent from TgAb has been brought up. In this study, we developed a liquid chromatography-tandem mass spectrometry (LC-MS/MS) method for Tg measurement and compared its clinical usefulness and analytical performance with immunoradiometric assay (IRMA) and chemiluminescence microparticle immunoassay (CMIA). Methods We developed trypsin digestion-peptide immunoaffinity LC-MS/MS methods for Tg measurement. Analysis was performed using an LC-MS/MS system consisting of Agilent 1290 LC (Agilent Technologies, Santa Clara, CA) and Qtrap 5500+ MS/MS system(ABSciex, Framingham, MA). Tryptic Tg specific peptide (FSPDDSAGASALLR) and cleavable internal standard peptide (LSFYWTTTGDPFFDSHSDSLL*RSGPYMPQ, L*: 13C6, 15N) were quantitated against 5-point linear calibration curve with Access Thyroglobulin Calibrators (Beckman Coulter, Pasadena, CA). Three levels of serum control were prepared (3.95, 12.5, and 39.4 ng/mL). Lower limit of quantitation (LLoQ), linearity, precision, and carryover were validated for the assay. For method comparison, serum from patients with DTC or Hashimoto thyroiditis was collected. Tg was also measured with IRMA using BRAHMS TG Plus (Thermo Fisher Scientific, Waltham, MA), and CMIA using Architect Thyroglobulin (Abbott laboratories, Illinois, CA). TgAb was measured with IRMA using BRAMS Anti-Tg (Thermo Fisher Scientific) and the cut-off for TgAb was 60 IU/mL. Also, interference of TgAb was evaluated by comparing the Tg recovery of five sets of TgAb negative samples spiked with diluents and high TgAb samples. Results Within-run and between-run imprecisions were CV 7.8%–13.0% and CV 8.9%–12.7%, respectively. LLoQ was 0.5 ng/mL. The assay was linear through 0.5–650 ng/mL. No significant carryover was observed within the analytical measurement range. Out of the total 193 samples, the number of samples above LLoQ in both LC-MS/MS, and IRMA or CMIA were 73 samples (45 TgAb-positive and 28 TgAb-negative samples) and 69 samples (42 TgAb-positive and 27 TgAb-negative samples), respectively. The result well correlated between LC-MS/MS and IRMA (R2 = 0.99), and LC-MS/MS and CMIA (R2 = 0.98) in overall samples. When analyzed according to TgAb status, LC-MS/MS and IRMA (R2 = 0.84), and LC-MS/MS and CMIA (R2 = 0.93) correlated rather poorly in TgAb-positive samples, while TgAb-negative samples were consistent (R2 = 0.99 for IRMA, R2 = 0.97 for CMIA). We found one case with systemic metastatic thyroid cancer that showed significantly high Tg levels in IRMA and LC-MS/MS (above the upper limit of quantitation), whereas below LLoQ in CMIA. The interference of TgAb was negligible in LC-MS/MS, while significant interference was observed in CMIA, with a mean recovery rate of 57.6% (4.3%–82.5%). In one sample, when measured with CMIA, thyroglobulin level was 3.29 ng/mL when mixed with diluent, but below LLoQ (0.14 ng/mL) when spiked with TgAb-positive samples (TgAb 1700 IU/mL). Conclusion The LC-MS/MS method for Tg measurement was less prone to antibody interference than immunoassays. Measurement of Tg using LC-MS/MS would be useful for reliable diagnosis and proper management in TgAb-positive patients.