UCLA

Many factors, including tumor size, plasma glucose level, and the time course of 2-[18F]fluoro-2-deoxy-D-glucose (FDG) in plasma can affect 18F-FDG uptake in tumors. The objective of this research work was to systematically investigate the effects of these biological factors on FDG uptake in tumors and to develop methodologies both to overcome measurement problems encountered and to account for the discovered effects to improve the value of FDG PET imaging.

In vitro tumor cell line studies were first performed using a PSAPD-coupled microfluidic cell culture chip (microfluidic beta camera). A novel medium infusion strategy--a switching strategy was used to extract the kinetic parameters of FDG uptake in tumor cells. Results showed that the switching strategy could provide reliable estimates of cellular FDG transport and uptake constants and the reliability of the model parameters can be further improved with an optimal design of fewer switching cycles. The experimental results also indicated that K1 (FDG transport constant into cells) and Ki (FDG uptake constant in cells) estimates of glioma cell line U87 and prostate cancer cell line CaP8 were inversely proportional to the glucose concentration in the medium, while the values of k2 (FDG transport constant out of cells), k3 (FDG phosphorylation rate constant in cells) and cellular glycolytic rate were not affected by the medium glucose level.

Cellular FDG uptake studies were also performed on multi-well cell culture plates to investigate the effects of medium glucose levels and free fatty acid levels on FDG uptake in cell lines of three different tumor types. It was found that FDG uptake in the tumor cell lines all decreased with increased medium glucose levels, but no effect was seen due to free fatty acid level changes. However, the relationships between FDG uptake and medium glucose level are different among different tumor types, i.e. the FDG uptake in mammary cell line MDA-MB-231 is not affected as much by medium glucose level over the normal physiological glucose range as those of U87 and PC3.

Based on the results from in vitro studies, a set of longitudinal quantitative mouse FDG-PET studies in tumor-bearing mice were launched to investigate whether the in-vitro findings are translatable to in vivo settings. Two tumor types, U87 and MDA-MB-231, were implanted in SCID mice. FDG time activity curves in plasma were derived from dynamic PET images and were used to estimate the tumor FDG kinetic parameters. A multivariate analysis was used to account for multiple factors (plasma glucose level, tumor size, tumor age) at the same time. The results show that the plasma glucose level affected tumor FDG uptake in ways consistent with the findings obtained in in-vitro experiments. Furthermore, the tumor FDG uptake constant was found to be stable for the two tumor types studied as tumor grew in size (before tumor necrotic core appeared). The study also demonstrated that quantitative longitudinal mouse FDG PET studies can be done in cancer research to provide valuable biological information.

In summary, new kinetic methods have been developed in this study to improve the quantitation of FDG uptake kinetics in tumors, both in-vitro and in-vivo. Findings from the quantitative studies have added new understandings of FDG uptake in tumors that have direct implication on improving the reliability of FDG PET studies in clinical settings.