Differences on metabolic behavior between intra and extrahepatic cholangiocarcinomas at 18 F-FDG–PET/CT: prognostic implication of metabolic parameters and tumor markers

Abstract

Background and purpose: Cholangiocarcinoma is an infrequent neoplasm barely studied with 18 F-FDG–PET/CT. We evaluated the metabolic behavior of cholangiocarcinoma in PET/CT according to its location (intra or extrahepatic) and analyzed the relationship between metabolic parameters of the primary tumor and tumor markers (CA19-9 and CEA), determining their prognostic significance. Methods: Retrospective study of PET/CT of 60 patients with untreated cholangiocarcinoma, divided into two groups according to tumor location. FDG uptake was evaluated visually and semiquantitatively [SUVmax and tumor-to-liver ratio (TLR)], and differences between intra and extrahepatic cholangiocarcinomas were tested, both for FDG uptake in the primary tumor and for the presence of regional or distant disease (per-patient), as well as regarding tumor marker levels. A correlation between metabolic parameters and tumor markers was performed, and prognostic value of these factors was determined (univariate and multivariate analyses). Results: Intrahepatic cholangiocarcinomas were significantly more FDG-avid than extrahepatic ones (p = 0.006 for SUVmax; p = 0.002 for TLR). There were differences neither between both groups considering the capacity of PET/CT to detect regional (p = 0.261) and distant involvement (p = 0.876), nor regarding the levels of tumor markers (p = 0.160 for CA19-9; p = 0.708 for CEA). Metabolic parameters and tumor markers showed a weak positive correlation (R 2 0.22–0.27). At the multivariate analysis, advanced stage (p = 0.024), increased CEA (p = 0.022), and higher TLR (p = 0.003) were significantly related with shorter overall survival. Conclusions: Intra and extrahepatic cholangiocarcinomas behave differently on PET/CT, though no differences between both groups exist in its capacity to detect regional or distant disease. Metabolic parameters and levels of tumor markers seem to relate with tumor burden, impacting in prognosis.