Clinical/methodical issue: The diagnostic work-up in patients with carcinoma of unknown primary (CUP) syndrome is extensive, highly time-consuming and cost-intensive and ultimately often fails to detect a primary site.
Standard radiological methods: In this context chest X-ray and computed tomography (CT) have been used as standard imaging modalities in CUP syndrome.
Methodological innovations: Since the introduction of positron emission tomography (PET) evaluation of tumor vitality has become possible. Furthermore, PET-CT hybrid scanners allow the combination of functional and morphological imaging.
Performance: Several meta-analyses have reported an additional overall detection rate between 24.5 % and 44 % by either PET or PET-CT. Metastatic localization (cervical versus extracervical) did not influence the performance. The sensitivity was usually high (> 80 %) but specificity was moderate ranging from 68 % to 88 % at best. If mentioned, the results obtained by fluorodeoxyglucose (FDG)-PET significantly changed the clinical management in approximately one third of the patients studied. In a direct comparison with PET alone, PET-CT did not depict significantly more primary tumors but was able to reduce false positive findings.
Achievements: To determine the real additional value of PET-CT in the diagnosis of CUP syndrome large prospective studies with more uniform inclusion criteria are needed. Despite the capabilities of FDG-PET-CT there is as yet no evidence that a potentially improved diagnostic algorithm is translated into a better patient outcome.
Practical recommendations: Nevertheless, FDG-PET-CT should be performed in all CUP patients where conventional imaging failed to detect a primary site or the results are equivocal. In CUP patients with cervical lymph node metastases PET-CT should be carried out prior to panendoscopy to reduce the number of false negative biopsies.