TY - JOUR T1 - <strong>Dynamic Modeling </strong><sup><strong>18</strong></sup><strong>F-FDG PET/CT for </strong><strong>H</strong><strong>epatic </strong><strong>M</strong><strong>etastasis in </strong><strong>Rabbit </strong><strong>VX2 </strong><strong>Tumor </strong><strong>Model and in Human</strong> JF - Journal of Nuclear Medicine JO - J Nucl Med SP - 243 LP - 243 VL - 59 IS - supplement 1 AU - Yongquan Huang AU - Dan Li AU - Zeqing Xu AU - Kaichen Huang AU - Huimin Lan AU - Hong Shan AU - Hongjun Jin Y1 - 2018/05/01 UR - http://jnm.snmjournals.org/content/59/supplement_1/243.abstract N2 - 243Objectives: The current clinical PET static scan method has disadvantages such as miscalculation, unable to distinguish the nonspecific lesions from hepatic metastasis, and management of overreliance on experience. PET dynamic scanning provides whole body organs of a full range of quantitative data, and leverages in tumor metabolism and dynamic distribution of PET tracers. There are some previous studies regarding carcinoma dynamic PET studies, but systematic dynamic modeling emphasizing on the hepatic metastasis was rarely found. The purpose of this study was to investigate the physiological process of 18F-FDG’s metabolism both in a VX2 model and in human, and to establish a protocol using dynamic modeling 18F-FDG PET/CT for hepatic metastasis, which will provide quantitative methodology to distinguish hepatic metastasis from the nonspecific lesions. Methods: Dynamic and static PET scans for 18F-FDG were performed on the rabbit VX2 tumor model and patients with hepatic metastasis. Biodistribution and metabolites of 18F-FDG in different region were quantified. The regions of interest (ROI) were draw from the original tumor and hepatic metastasis along with reference regions (lung, heart, liver and kidney). The time-activity curve (TAC) was acquired from the dynamic scans for these ROIs. The radioactivity calculated by averaging the whole voxel’s values within the ROIs were carried out in all analysis. The dynamic modeling including 3-tissue-compartment model (3TCM) and Patlak plot from the rabbits and the patients were performed in Matlab for kinetics parameter estimation. Results: Three rabbits of VX2 tumor model and 10 hepatic patients with variable hepatic metastasis were detected (Supporting-information Figure 1). All lesions were identified by focal 18F-FDG accumulation and SUVmax or SUVmean than in nontumor tissues. There were visual trend and quantitative differences between original tumor and hepatic metastasis TAC curves. Tumor SUVmax and SUVmean exceeded healthy liver levels in 83% and 55%. Weighted least squares principle was used to fit the 3TCM and Patlak models. Parameters for 3TCM and Patlak models were analyzed and compared. Conclusion: Both 3TCM and Patlak models have fitted well with the original TAC data from dynamic PET scans. The dynamic models have proved to be helpful of quantitative discriminating hepatic metastasis from nonspecific metabolites. ER -