MSACL 2022 Abstract

Background
Measurement of thyroglobulin (Tg) in serum and plasma is used to monitor patients after treatment for differentiated thyroid carcinoma (DTC). Anti-Tg autoantibodies (Tg-AAb) can interfere with Tg measurements and may cause false-low results in Tg immunoassays (Tg-IAs).. Digestion of samples followed by quantitative analysis of Tg-specific peptides using liquid chromatography–tandem mass spectrometry (LC-MS/MS) has demonstrated the ability to overcome interference from endogenous Tg-AAb, which are present in approximately 20% of patients. LC–MS/MS assays for thyroglobulin have been developed which have limits of quantitation (LOQ) between 0.2 and 0.5 ng/mL, while typical LOQ of commercial Tg-IAs is 0.1 ng/mL. Our goal was to develop an LC-MS/MS method for Tg using conventional flow rate LC separation with sensitivity comparable to commercial Tg-IAs and to assess the method’s performance using a novel instrument, Sciex Triple Quad 7500.

Methods
Study samples were residual, deidentified serum submitted for routine testing to ARUP Laboratories (n=540) and serum quality control samples (n=162). Sample preparation was performed as follows: Tg was enriched from human serum samples, internal standard (stable isotope-labeled winged peptide) was added to the samples, proteins were denatured, reduced, and digested using trypsin. A Tg-specific tryptic signature peptide (VIFDANAPVAVR) and the internal standard were enriched from the digests using immunoaffinity enrichment. The samples were injected onto the two-dimensional LC system either coupled to a Sciex Triple Quad 6500 or Sciex Triple Quad 7500 and the instruments’ performance compared. Instrumental analysis time was 6.5 minutes per sample.

Results
Total imprecision of replicate analysis of prepared serum samples containing 0.07, 0.10, 0.74, 13.8, and 255 ng/mL of Tg was 13.3%, 19.3%, 11.9%, 13.8% and 12.3%, respectively. LOQ of Tg in serum samples was 0.07 ng/mL (105 amol/mL of the Tg dimer; 23 amol injected on column), and limit of detection (LOD) was 0.03 ng/mL. Linear dynamic range of the assay was comparable between the Sciex 7500 and 6500 instruments (upper limit of linearity of 400 ng/mL). Signal-to-noise ratio for the standard of the signature peptide (100 zmol injected on column, n=3) was 259. Comparison of Tg concentrations using the two evaluated MS platforms showed good agreement between the methods (all samples, concentrations range 0-400 ng/mL, n=702): LC-MS/MS7500 = 1.01* LC-MS/MS6500 - 0.00, r=0.999; samples with Tg <2.0 ng/mL, n=424: LC-MS/MS7500 = 0.95* LC-MS/MS6500 - 0.02, r=0.977). Absolute peak area of internal standard was 20.8-fold higher using the Sciex 7500 than the Sciex 6500 instrument. Enhancement in the signal-to-noise ratio was concentration-dependent; mean values for the increase in the signal-to-noise ratio at concentrations <0.2, 0.21-5.0; 5.1-20; 20.1-100 and >100 ng/mL were 3.6, 4.9, 3.9, 2.9, and 1.9, respectively. In analyses performed on the Sciex 7500, Tg concentrations >0.07 ng/mL were observed in 40% of the samples with undetectable Tg by commercial immunoassay (Beckman Coulter, Brea, CA; LOQ 0.1 ng/mL) and an LC-MS/MS method with LOQ 0.5 ng/mL.

Conclusions
We developed a LC-MS/MS method for thyroglobulin with instrumental analysis performed on a Sciex Triple Quad 7500 instrument, evaluated the method’s performance, and observed significant enhancement in the sensitivity as compared to the Sciex 6500. The higher instrument sensitivity may allow for improvements in care for patients monitored for DTC recurrence. The developed method offers sensitivity that was previously only found with microflow-based methods and may offer a comparative advantage in robustness.