Abstract
BACKGROUND: Dynamic course of flourine-18 fluorodeoxyglucose positron emission tomography/computed tomography ((18)F-FDG PET/CT) Patlak muti-parametric imaging spatial distribution in the targeted tissues may reveal highly useful clinical information about the tissue's metabolic properties. The characteristics of the Patlak multi-parametric imaging in lung cancer and the influence of different delineation methods on quantitative parameters may provide reference for the clinical application of this new technology. METHODS: A total of 27 patients with pathologically diagnosed lung cancer underwent whole-body dynamic (18)F-FDG PET/CT examination before treatment. Parametric images of metabolic rate of FDG (MR(FDG)) and Patlak intercept (or distribution volume; DV) were generated using Patlak reconstruction. The values of primary lung cancer lesions, target-to-background ratio (TBR), and contrast-to-noise ratio (CNR) were investigated using contour delineation and boundary delineation. Statistical analysis was performed to analyze the relationship between multi-parametric images and clinicopathological features, and to compare the effects of contour delineation and boundary delineation on quantitative parameters. RESULTS: MR(FDG) images showed higher TBR and CNR than did standardized uptake value (SUV) images. There were significant differences in MR(FDG-max), MR(FDG-mean), and MR(FDG-peak) among groups with different tumor diameters and pathology types (P<0.05). Moreover, the metabolic parameters of MR(FDG) were higher in patients with tumor diameters ≥3 cm and squamous carcinoma. The differences of the maximum and peak values of MR(FDG) and DV were not statistically significant in the different outlining method subgroups (all P>0.05). However, the difference of the mean values of MR(FDG) and DV were statistically significant in the different outline method groupings (all P<0.05). CONCLUSIONS: Dynamic (18)F-FDG PET/CT Patlak multi-parametric imaging can obtain quantitative values for lung cancer with high TBR and CNR. Moreover, the multi-parameters are various from different pathology types to tumor size. Different delineation methods have a greater influence on the mean value of quantitative parameters.