Objective: To assess the diagnostic utility of [(18)F]flu-orodeoxyglucose positron emission tomography (FDG PET) in the follow-up of patients with differentiated thyroid cancer (DTC).
Methods: In this study, we used strict definitions of presence and absence of the disease and performed all FDG PET scans while the patients were not taking levothyroxine (LT4). We report the results of conventional FDG PET scans obtained during the follow-up of 50 nonselected patients with DTC (34 female and 16 male patients; median age, 40.5 years; range, 18 to 68). All FDG PET scans and measurement of thyroglobulin (Tg) levels were performed while the patients were not taking LT4 (thyrotropin>or=25 microIU/mL). Tg antibodies were negative in all patients. The initial surgical procedure was total thyroidectomy in all cases, and 26 patients underwent additional operations (2 to 4 procedures). Radioactive iodine (131I) therapy was given to 48 patients (median dose, 5,550 MBq). In 42 patients, FDG PET was used for evaluation of Tg-positive (Tg>2 ng/mL in the absence of LT4 therapy), scan-negative disease. In 8 patients, Tg was <or=2 ng/mL but other findings suggested the presence of the disease (detectable Tg in 7 cases and abnormal ultrasound findings in 3). Disease was considered present if confirmed by fine-needle aspiration, histopathologic examination of subsequent surgical specimens, or persistent elevation of Tg levels (>2 ng/mL without LT4 therapy) for >1 year. Disease was considered absent if Tg was <or=2 ng/mL and at least all the following imaging studies were negative: diagnostic radioiodine whole-body scan, chest radiography or spiral computed tomography of the chest, and high-resolution neck ultrasonography. If all these criteria were present, a positive FDG PET scan was considered a false-positive study.
Results: FDG PET scans were positive in 27 patients (54%) and negative in 23 (46%). FDG PET results were true-positive in 26 cases, false-positive in 1, true-negative in 7, and false-negative in 16. The sensitivity, specificity, and positive and negative predictive values were 61.9%, 87.5%, 96.3%, and 30.4%, respectively.
Conclusion: FDG PET scanning is moderately sensitive and specific for detection of persistent or recurrent DTC.