Issue: The use of thyroglobulin (Tg) as tumor marker in differentiated thyroid cancer (DTC) is limited in the presence of thyroglobulin-antibodies (TgAb) but it is generally believed that this is true only for TgAb concentrations over the normal cut off point.
The aim: Of this study was to investigate if TgAb-s in the normal range, considered to be physiological, may also influence the accuracy and clinical relevance of Tg measurement.
Methods: Recombinant human TgAb (Roche) was added stepwise to serum-samples (n=45) with TgAb concentrations near to the analytical sensitivity of the method (10 IU/ml), aiming to have TgAb concentrations of 50100150 and 200 IU/ml (ECLIA Elecsys 2010 Roche, normal cut off <115 IU/ml). After this, Tg levels were measured at all TgAb concentrations by electrochemiluminescence immunoassay (ECLMA, Elecsys 2010, Roche). Additionally, 134 samples from 27 patients with DTC were measured for Tg, Tg-recovery (Tg%) and TgAb.
Results: In the in vitro experiment, TgAb and Tg concentrations showed strong correlation (r=0.93, P<0.01) both at normal and elevated TgAb levels, which could be described mathematically as: Loss of Tg=−0.43 Ln(TgAb IU/ml)+1,06. Patients with non-detectable Tg had higher antibody levels than those with detectable Tg. There was a rather weak negative correlation (r=−0,32 P<0.001) of Tg% to TgAb and in 19% of the samples the results were clinically discordant. In 2/27 patients, on-T4 Tg levels of <2.0 ng/ml were corrected to be >2,0 ng/ml by using the above function. Subsequent off-T4 Tg levels appeared to be significantly elevated in both.
Conclusion: Physiological (normal) TgAb concentrations may also decrease serum Tg but their effect can be calculated from the actual Tg and TgAb concentrations by the mathematical model described. The findings stress the importance of parallel Tg and TgAb measurements in patients with DTC expected to have undetectable or low Tg.