A-047 Evaluation of the Clinical Performance of the Thyroglobulin Assay on the Atellica IM Analyzer

Abstract Background Thyroglobulin (Tg) is a protein produced by thyroid follicular cells and functions as a prohormone in the intra-thyroid synthesis of thyroxine (T4) and triiodothyronine (T3). Lysosomes containing proteases cleave T4 and T3 from Tg, resulting in T4 and T3 release. Tg is present in the serum of healthy individuals and can be elevated in a number of disorders such as subacute thyroiditis, autonomous adenoma, amiodarone induced thyrotoxicosis and thyroid carcinoma. An important application of Tg measurement is the post-operative follow-up of patients with differentiated thyroid cancer (DTC). Quantitative Tg measurements are used as an aid in monitoring DTC patients who have undergone thyroidectomy, with or without radioiodine ablation. Here, studies were performed to evaluate reproducibility of the Atellica IM Tg assay on the Atellica IM analyzer and to evaluate assay clinical performance at a cutoff of 0.2 ng/mL. Values >0.2 ng/mL were deemed “positive” (evidence of possible cancer recurrence) as defined in 2015 ATA Guidelines. Methods Reproducibility testing was performed per CLSI EP05-A3, involving three levels of quality control (QC) material (0.221, 1.98, 25.8 ng/mL; 0.335, 3.00, 39.1 pmol/L) and a five-level patient sample pool (4.36, 44.0, 74.5, 120, 129 ng/mL; 6.61, 66.7, 113, 182, 195 pmol/L) tested in two runs per day for 5 days, three sites, three Atellica IM analyzers, and three Atellica IM Tg reagent lots. The same three reagent lots were tested at each of the three sites. Samples for clinical performance evaluation were prospectively collected from eligible subjects at sites in the United States. Testing was completed at one internal site. Primary analysis included 291 evaluable samples from 189 unique subjects. Assay result was compared to the status of structural disease as determined by cross-sectional or functional imaging results collected within 30 days of study blood draw. Clinical sensitivity, specificity, PPV and NPV were computed using a bootstrapping method to account for repeated Tg measurements within subjects. Results Quality control and patient sample pools recovered mean values between 0.221 and 129 ng/mL (0.335 and 195 pmol/L), spanning the assay range of 0.05-150 ng/mL (0.076-227.250 pmol/L) and covered the medical decision level of 0.2 ng/mL (0.3 pmol/L). Coefficients of Variation (CV) for reproducibility were =5.0% for samples recovering =1.000 ng/mL (1.5 pmol/L) and 9.0% for sample recovering <1.000 ng/mL. Sensitivity was 98.2% (95%CI 94.6%, 100.0%), specificity 53.4% (47.8%, 58.0%), PPV 10.0% (8.7%, 11.2%), and NPV 99.8% (99.5%, 100.0%) when assay result was compared to status of structural disease. Conclusion Overall, the Atellica IM Tg assay demonstrated acceptable clinical performance as an aid in the monitoring of DTC thyroidectomy patients with or without radioiodine ablation.