Thyroid cancer recurrence in patients clinically free of disease with undetectable or very low serum thyroglobulin values.

DESIGN This was a retrospective clinical study. SETTING The study was conducted at a university-based tertiary cancer hospital. PATIENTS One hundred seven patients had initial thyroid cancer surgery and subsequent remnant radioiodine ablation. Patients underwent recombinant human TSH (rhTSH)-mediated diagnostic whole-body scan and rhTSH-stimulated thyroglobulin (Tg) measurement before April 2001 if they had no antithyroglobulin antibodies, were clinically free of disease, and had one or more undetectable (≤0.5 ng/ml) or low (0.6-1 ng/ml) basal Tg measurements on levothyroxine. Patients were stratified according to their rhTSH-Tg responses: group 1, Tg 0.5 ng/ml or less (68 patients); group 2, Tg from 0.6 to 2.0 ng/ml (19 patients); and group 3, Tg greater than 2 ng/ml (20 patients). MAIN OUTCOME MEASURES Tumor recurrence was measured. RESULTS In group 1, two of 62 patients (3%) with follow-up recurred. In group 2, 63% converted to group 1, whereas two of 19 (11%) converted to group 3 and then recurred. Sixteen of the initial 20 group 3 patients (80%) recurred, including recurrence rates of 69 and 100% for those with an initial rhTSH-Tg greater than 2.0 ng/ml but 5.0 ng/ml or less, and 4.6 ng/ml or greater, respectively. One group 3 patient died of distant metastases. rhTSH-Tg more accurately predicted tumor recurrence than basal Tg. An rhTSH-Tg threshold of 2.5 ng/ml or greater optimally predicted future recurrence with sensitivity, specificity, and negative and positive predictive values of 80, 97, 95, and 84%, respectively. CONCLUSIONS The prevalence of postablation thyroid cancer recurrence is predicted by the rhTSH-Tg response with an optimal Tg threshold of 2.5 ng/ml. Still, recurrent disease occurs in some patients with an initial rhTSH-Tg of 0.5 ng/ml or less.