Patlak Reconstruction Using Dynamic ¹⁸F-FDG PET Imaging for Evaluation of Malignant Liver Tumors: A Comparison of HCC, ICC, and Metastatic Liver Tumors

Purpose of the Report: The aim of this study was to explore the different patterns of dynamic whole-body (D-WB) FDG PET/CT parameters among liver malignancy types as potential diagnostic clues and investigate the association between static and dynamic PET/CT parameters for each tumor histology.Patients and Methods: Seventy-one patients with intrahepatic cholangiocarcinoma (ICC), metastatic liver tumor (MLT), or hepatocellular carcinoma (HCC) who underwent D-WB and static dual-time-point FDG PET/CT were enrolled. We obtained Pearson correlation coefficients between the metabolic rate of FDG (MR (FDG) ; mg/min/ 100ml) or distribution volume of free FDG (DV (FDG) , %) and static PET/CT parameters. We compared MR (FDG) and DV (FDG) values by tumor type and performed receiver operating characteristic analyses for MR (FDG) and static images.Results: A total of 12 ICC, 116 MLT, and 36 HCC lesions were analyzed. MR (FDG) and DV (FDG) showed excellent correlation with early (SUV (e) ) and delayed SUV (max) (SUV (d) ) ( r = 0.71 similar to 0.97), but DV (FDG) in the HCC lesions did not ( r = 0.62 and 0.69 for SUV (e) and SUV (d) , respectively) ( P < 0.001 for all). HCC lesions showed significantly lower MR (FDG) (2.43 +/- 1.98) and DV (FDG) (139.95 +/- 62.58) than ICC (5.02 +/- 3.56, 207.06 +/- 97.13) and MLT lesions (4.51 +/- 2.47, 180.13 +/- 75.58) ( P < 0.01 for all). The optimal MR (FDG) could differentiate HCC from ICC and MLT with areas under the curve of 0.84 and 0.80, respectively. Metastatic liver tumor lesions showed the widest distribution of MR (FDG) and DV (FDG) values but with no significant difference among most primary sites.Conclusions: MR (FDG) was strongly correlated with SUV (max) in the 3 malignancies and showed utility for differentiating HCC from ICC and MLT. Each tumor type has a different glucose metabolism, and D-WB FDG PET/CT imaging has the potential to visualize those differences.