Endocrine Abstracts

Introduction

Up by 30% of patients have recurrence in the first years after treatment of differentiated thyroid cancer (DTC) so it is necessary to have highly sensitive assays that allow adequate long-term follow-up. Nowadays, the measurement of serum thyroglobulin (Tg) is used as a tumor marker to monitoring DTC and must be determined along with levels of antithyroglobulin antibodies, since they can interfere in Tg immunoassays. Therefore, it is recommended to determine both using the same method and equipment.

Objective

The aim of this study was to evaluate the interchangeability between the Tg-ultrasensitive results using two types of immunoassays.

Materials and methods

Parallel determination of serum Tg concentration in 80 samples of patients with thyroid pathology, using the analyzers Modular E-170® (electrochemiluminescence; Roche-Diagnostics) and Architect i4000SR® (chemiluminescent-microparticle-CMIA; Abbott-Diagnostics). The statistical analysis has been developed with the Method-Validator software (v1.19), following the recommendations of the Spanish Society of Laboratory Medicine (SEQCML) for comparison of measurement procedures, performing the analysis of the agreement by Bland-Altman plot and regression by the methods of Deming and Passing-Bablok.

Results

Conclusions

By the data obtained, both methods would be interchangeable according to Bland–Altman analysis (the value 0 is included within CI) and Deming’s method (value 0 for the ordinate at the origin and the value 1 for the slope are within their respective CI). As well as the regression analysis can be performed (r≥0.975). However, Passing-Bablok method indicated the presence of a systematic proportional difference since the value 1 is not within the CI for the slope, although it is very close. This type of error could be due to the fact that both assays have different methodology in their immunoassays and could be associated with the presence of different interferences described in the literature (Streptavidin/Biotin technology), thus the patients should always be tested with the same highly sensitive assay. It would be necessary to increase the number of samples in order to cover the dynamic range of both techniques. Although this minimal difference has no clinical relevance, it is a priority to guarantee highly accurate results at the detection limit, in order to detect tumor recurrence and/or persistence early.