FDG-PET in medullary thyroid cancer: A disappointment?

1271 Objectives: Early detection of residual or remote tumor sites of medullary thyroid cancer (MTC) is a diagnostic challenge. Conflicting data have been published on the value of FDG-PET in this malignancy. The present two-center study evaluates the impact of FDG-PET and PET/CT on lesion detection and management of patients with MTC. Methods: Twenty-nine MTC patients [14F, 15M, age 12-71 yrs] underwent 45 whole body FDG studies [21 PET and 24 PET/CT]. Scans were performed at staging [n=6] and in search of tumor recurrence [n=39], following elevation of calcitonin or suspicious findings on morphological imaging. Results of blinded evaluation of PET and PET/CT were compared with US, CT, and surgery, when available, and with clinical follow-up and serial calcitonin level measurements. Results: Twelve out of the 45 FDG scans [PET-2, PET/CT-10] showed no abnormal FDG avidity, despite elevated calcitonin in 9 of these 12 studies. FDG-avid sites were visualized in 33 studies [PET-19, PET/CT-14]. In a subgroup of 14 studies, followed by surgery, FDG imaging diagnosed disease in twelve histologically confirmed sites [9 neck, 3 mediastinum]. In 5/14 studies, FDG imaging failed to detect surgically proven tumor sites in the neck, that were diagnosed by US prior to intervention in 3 cases. Also, in 5/14 studies, tumor involvement on CT and/or US was more extensive than on PET, again with verification at surgery. A similar pattern was found in the additional 31 FDG scans, with as yet no histological verification. More tumor sites were indicated in the neck [n=12] and in the mediastinum [n=8] on US/CT than on PET [n=8 and n=7, respectively]. Other remote sites, including lung, liver, adrenal and bone, followed the same pattern [15 FDG-avid sites vs. 24 sites on CT]. Only one PET/CT study revealed two previously unrecognized tumor sites [neck-1, spine-1], that were subsequently confirmed by CT. However, in the subgroup of 21 PET-only studies, 12 FDG-avid sites were suspicious for malignancy with negative CT in 12/12 and negative surgical exploration in 3/12 sites. Overall, FDG-PET results improved patient management in 2/45 studies, by guiding re-exploration of cervical and mediastinal involvement. Conclusions: Our study suggests that in patients with MTC, FDG-PET is less sensitive than US and CT for assessment of disease in the neck and mediastinum, notwithstanding improved specificity of PET/CT over PET-alone.