Utility of fluorodeoxyglucose-PET in patients with differentiated thyroid carcinoma

In differentiated thyroid carcinoma, persistent plasma thyroglobulin (Tg) is a specific marker for persistent or recurrent disease after thyroidectomy and radioiodine ablation. When Tg remains elevated and no substrate can be found on whole-body radioiodine imaging (131I-WBS), or even when recurrent disease is suspected with normal Tg, conventional imaging (CI) is often insufficient. As fluorodeoxyglucose (FDG)-PET has proven to be an effective modality for detecting various types of cancer, the utility of FDG-PET was analysed and compared with CI in this retrospective study in patients with differentiated thyroid cancer.

After total thyroidectomy and radioiodine ablation, 68 FDG-PET scans were performed in 39 patients with elevated Tg levels or clinical suspicion of recurrent disease. At the time of FDG-PET, 54 131I-WBS (in 30 patients) were negative, 14 (in 11 patients) were equivocal. Tg was normal at the time of 14 scans (10 patients) and elevated in 54 (in 33 patients). FDG-PET results were compared with histology, 131I-WBS and CI and clinical follow-up. Sensitivity and specificity were evaluated in various subgroups.

Overall, there were 35 true-positive, two false-positive, 20 true-negative and three false-negative FDG-PET scans. In six of these cases (one true positive, five true negative) FDG-PET was repeated without intervention and in an additional eight FDG-PET scans no definite conformation of abnormal FDG-PET could be obtained, so these results were not used for statistical analysis. Sensitivity, specificity, PPV and NPV for the whole group were 92, 88, 94 and 83%, respectively. In 38 scans performed on 31 patients with elevated Tg levels, who were not known with recurrence, this was 84, 100, 100 and 75%, respectively. In 16 scans in 10 patients with known recurrence (all with elevated Tg), sensitivity and PPV were 100% without false-positive or false-negative results. When Tg was not detectable (14 scans in 10 patients), sensitivity, specificity, PPV and NPV were 100, 75, 60 and 100%, respectively. After 35 FDG-PET scans (51%), there was a change in patient management by avoiding ineffective 131I treatment, by guiding surgical reintervention, or avoiding futile surgery. One FP FDG-PET resulted in an unnecessary surgical procedure. In 33 cases, FDG-PET did not lead to a change in treatment policy, which retrospectively would have been beneficial in six cases.

FDG-PET affected patient management in patients with differentiated thyroid cancer and negative 131I-WBS, not only when Tg is elevated, but also when Tg is not detectable and therefore the use of FDG-PET as a diagnostic tool is justified in these patients.