Serum thyroglobulin concentrations predict disease-free remission and death in differentiated thyroid carcinoma.

OBJECTIVE Most studies on the diagnostic value of serum thyroglobulin (Tg) concentrations in differentiated thyroid carcinoma (DTC) use fixed cut-off levels in heterogeneous groups of patients with respect to initial therapy and do not provide prognostic data. The objective was to investigate the prognostic values of serum Tg for disease-free remission and death, measured at fixed time-points after initial therapy using receiver operator characteristic (ROC) curve analyses. DESIGN Single-centre observational study with 366 consecutive patients with DTC, who had all been treated according to the same protocol for initial therapy and follow-up. METHODS Tg concentrations were measured at five fixed time-points after initial surgery. Tg cut-off values with the highest accuracy were calculated with ROC analyses. RESULTS During the 8.3 +/- 4.6 years of follow-up, 84% of the patients were cured. Pre-ablative Tg levels were an independent prognostic indicator for disease-free remission (Tg cut-off value 27.5 microg/l, positive predictive value 98%). The highest diagnostic accuracies of serum Tg for tumour presence were found during TSH-stimulated Tg measurements, 6 months after initial therapy (Tg cut-off value 10 microg/l; sensitivity 100%, specificity 93%). DTC-related mortality was 14%. TSH-stimulated Tg levels before ablation and 6 months after initial therapy were independent prognostic indicators for death. CONCLUSION Optimal institutional Tg cut-off levels for diagnosis and prognosis should be defined using ROC analyses for each condition and time-point. Tg measurements 6 months after initial therapy during TSH stimulation had an excellent diagnostic value. Tg levels are independent prognostic indicators for disease-free remission and death. Using this strategy, high-risk patient groups can be selected based on Tg levels, in addition to conventionally used prognostic indicators.