Abstract
1007
Background: While targeting somatostatin receptors (SSTR) is continuing to gain ground for image-guided treatment of various human malignancies, that approach is only effective in roughly 40% of patients with neuroendocrine tumors (NETs). As such, alternative strategies applicable to a broader range of patients are required. [18F]fluoro-L-DOPA (FDOPA) is a PET radiotracer currently in clinical trials for imaging of patients with NETs. The diagnostic utility of this agent, however, is limited by significant off-target retention in the pancreas, liver, urinary excretion system, gallbladder, and biliary tract. That non-specific retention is in part attributed to the metabolism of FDOPA, which reduces signal-to-noise ratios and dampens disease delineation. While enzyme inhibitors such as carbidopa can aid in reducing that metabolism, they can also block radiotracer uptake in tumors. More effective strategies in reducing FDOPA metabolism can significantly improve its utility for PET imaging of NETs. Enzymes in the dopaminergic pathway typically exhibit a high degree of specificity for individual amino acid enantiomers. That includes the enzyme L-amino acid decarboxylase (AADC), which metabolizes L-DOPA to dopamine. Given that AADC does not recognize, and therefore cannot metabolize D-DOPA, we assessed [18F]fluoro-D-DOPA for PET imaging of NETs in a side-by-side comparative study with FDOPA.
Methods: Six human NET cell lines, including intestinal and colorectal cancers, insulinomas and pheochromocytomas, were selected. Those cell lines were assessed for LAT1 (an amino acid transporter involved in FDOPA uptake) gene and protein expression, by flow cytometry and RT-PCR. Radiotracers were synthesized using a Trasis AllinOne synthesizer and subsequently analyzed by HPLC for radiochemical purity and enantiomeric excess. Both radiotracers were evaluated in vitro in all six cell lines prior to in vivo evaluation in mice bearing human pheochromocytomas, colorectal and intestinal NET xenografts. Animal studies entailed PET-CT imaging, ex vivo biodistribution and immunohistochemical analyses of excised tumor tissues.
Results: LAT1 protein expression was observed in all cell lines with intestinal NET cells having the lowest and pheochromocytoma cells having the highest expression. LAT1 gene expression was also highest in pheochromocytoma cells, but lowest in colorectal cancer cells. FDOPA was prepared in 70 mins with an average non-decay corrected radiochemical yield of 40±10% (n=15). [18F]Fluoro-D-DOPA was synthesized in 90 mins with an average non-decay corrected radiochemical yield of 18±5% (n=15). In both cases, radiochemical purities and enantiomeric excess were > 97%. Both radiotracer formulations were stable and did not show any signs of radiolysis or degradation (tested up to 9 hrs post synthesis). In vitro binding assays showed uptake of both radiotracers in all NET cell lines. LAT1 inhibition resulted in a > 2-fold reduction in radiotracer uptake, indicating a LAT1-mediated mechanism. However, under the evaluated incubation conditions, FDOPA binding was consistently higher than that of [18F]fluoro-D-DOPA in all cell lines reaching 25±5% (versus 5±2% for [18F]fluoro-D-DOPA) in pheochromocytoma cells. Interestingly, while in vivo PET-CT images also showed higher tumor uptake of FDOPA in all tumor types, [18F]fluoro-D-DOPA had better overall clearance from non-specific tissues including pancreas and liver.
Conclusions: Side-by-side comparison of both [18F]fluoro-DOPA enantiomers demonstrated key advantages and pitfalls of [18F]fluoro-D-DOPA over the clinical radiotracer FDOPA for NET PET imaging applications. While much remains to be done, this study highlights the capabilities of [18F]fluoro-D-DOPA for detecting mid-gut NETs that are more difficult to delineate with FDOPA. Funding: This work was supported by the Neuroendocrine Tumor Research Foundation and The Education and Research Foundation for Nuclear Medicine and Molecular Imaging.