Abstract

Objective: To describe the diagnostic accuracy of PET-CT with ¹⁸F-FDG for the location of primary tumors of unknown origin.

Methods: Patients with a clinical diagnosis of cancer of unknown primary origin who underwent a PET-CT study with ¹⁸F-FDG were included. Cases of primary cancer already localized and cases that had already received treatment were excluded. The detection rate (DR) and the proportion of false positives (FP) were calculated.

Results: One-hundred patients with a clinical diagnosis of cancer of unknown primary origin who underwent ¹⁸F-FDG PET-CT were studied; in 74 cases there was at least one positive finding by PET-CT, of which a malignant etiology was confirmed by biopsy or follow-up in 47; in 4 the etiology was benign / premalignant and in 5 patients the cause of uptake was determined as infectious / inflammatory. The DR obtained was 47%, and the FP proportion was 27%. In patients with malignancies confirmed by biopsy the most frequent diagnosis was poorly differentiated adenocarcinoma, being the most common site of location the lung, followed by the pancreas.

Conclusions: PET-CT with ¹⁸F-FDG is useful in the detection of tumors of unknown primary origin; however, this method should be used when conventional imaging methods were not sufficient to determine the primary site of origin.

Key words: PET-CT, positron emission tomography, unknown primary tumors, occult primary.