Serum calcitonin and carcinoembryonic antigen (CEA) are markers of recurrent or persistent disease in medullary thyroid cancer (MTC). However, conventional imaging often fails to localize metastatic disease. Our aim was to compare fluorine-labeled dihydroxyphenylalanine ((18)F-DOPA) and (18)F-FDG PET/CT with multidetector CT (MDCT) and MRI in recurrent or persistent MTC.
Methods: Nineteen MTC patients with increased calcitonin or CEA on follow-up (mean ± SD, 93 ± 91 mo; range, 4-300 mo) after primary therapy were prospectively imaged with 4 techniques: (18)F-DOPA PET/CT, (18)F-FDG PET/CT, MDCT, and MRI. Images were analyzed for pathologic lesions, which were surgically removed when possible. The correlation between the detection rate for each method and the calcitonin and CEA concentrations and histopathologic findings was investigated.
Results: On the basis of histology and follow-up, one or more imaging methods accurately localized metastatic disease in 12 (63%) of 19 patients. The corresponding figures for (18)F-DOPA PET/CT, (18)F-FDG PET/CT, MDCT, and MRI were 11 (58%) of 19, 10 (53%) of 19, 9 (47%) of 19, and 10 (59%) of 17, respectively. Calcitonin and CEA correlated with (18)F-DOPA PET/CT (P = 0.0007 and P = 0.0263, respectively) and (18)F-FDG PET/CT findings (both P < 0.0001). In patients with an unstable calcitonin doubling time (n = 8), (18)F-DOPA and (18)F-FDG PET/CT were equally sensitive. In contrast, for patients with an unstable CEA doubling time (n = 4), (18)F-FDG PET/CT was more accurate.
Conclusion: For most MTC patients with occult disease, (18)F-DOPA PET/CT accurately detects metastases. In patients with an unstable calcitonin level, (18)F-DOPA PET/CT and (18)F-FDG PET/CT are complementary. For patients with an unstable CEA doubling time, (18)F-FDG PET/CT may be more feasible. MRI is sensitive but has the highest rate of false-positive results.