Detection of occult primary tumors in patients with cervical metastases of unknown primary tumors: comparison of (18)F FDG PET/CT with contrast-enhanced CT or CT/MR imaging-prospective study.

PURPOSE: To assess diagnostic accuracy of fluorine 18 ((18)F) fluorodeoxyglucose (FDG) combined positron emission tomography (PET) and computed tomography (CT) in the detection of occult primary tumors and determination of optimal care in patients with cervical metastasis of an unknown primary tumor (CUP) compared with contrast material-enhanced CT alone or combined contrast-enhanced CT and magnetic resonance (MR) imaging (CT/MR imaging).

MATERIALS AND METHODS: Institutional review board approval and written informed consent were obtained. In total, 56 patients with initially undetected tumors after endoscopic or physical examination were prospectively assessed with (18)F FDG PET/CT and contrast-enhanced CT or contrast-enhanced CT/MR imaging. The contrast-enhanced CT/MR images were read in combination. Results of guided biopsy with general anesthesia served as the reference standard. Diagnostic values of (18)F FDG PET/CT, contrast-enhanced CT, and contrast-enhanced CT/MR imaging were compared with the McNemar test.

RESULTS: Primary tumors were detected at 32 sites in 31 (55%) of 56 patients. There were 26 tumors in the palatine tonsil, two in the hypopharynx, two in the base of the tongue, and two in the nasopharynx. PET/CT depicted 22 (69%) of 32 primary tumors, but it failed to depict primary tumors in 10 (31%) of 32 cases. Overall, sensitivity of PET/CT (69%) in detection of primary tumors was higher than that of contrast-enhanced CT (16%) (P < .001) or contrast-enhanced CT/MR imaging (41%) (P = .039), while specificity of these methods did not differ (88%, 76%, and 59% for PET/CT, contrast-enhanced CT, and contrast-enhanced CT/MR imaging, respectively; P > .4). Diagnostic performance (area under the receiver operating characteristics curve [AUC] = 0.759) of PET/CT in tumor detection was significantly better than that of contrast-enhanced CT alone (AUC = 0.531) (P = .001) or contrast-enhanced CT/MR imaging (AUC = 0.537) (P = .036). PET/CT depicted primary tumors in eight (50%) of 16 cases of false-negative CT/MR imaging findings, one distant metastatic case, and two cases of synchronous cancer.

CONCLUSION: (18)F FDG PET/CT is more sensitive in detection of primary tumors than is contrast-enhanced CT or contrast-enhanced CT/MR imaging in patients with CUP; therefore, it may lead to improved therapeutic planning in these patients.