Abstract:
This work details the investigation of new approaches for in vivo diabetes imaging, including the optimization of the required radiochemical syntheses.
Diabetes mellitus is a group of metabolic disorders characterized by impaired glucose homeostasis that arises from the loss or improper functioning of insulin-producing beta cells (β cells). Despite significant advances in diabetes medicine, the changes that occur to the number of viable β cells, the β cell mass (BCM), over the course of the illness, are poorly understood in both forms of the disease. This is in large part due to the lack of an effective non-invasive method for localizing and quantifying the BCM in vivo. The development of such a method could potentially pave the way for a paradigm shift in diabetological research, as it would allow for the longitudinal tracking of the BCM in affected patients as the disease progresses. Nuclear medical imaging techniques, such as positron emission tomography (PET), have come to the forefront of this development; however, these methods rely on the development of highly specific radiotracers that target β cell-specific biomarkers.
The glucagon-like peptide-1 receptor (GLP-1R) is an excellent biomarker for in vivo β cell quantification. Radiolabeled derivatives of the GLP-1R orthosteric agonist exendin have been developed as specific GLP-1R targeting radiotracers; however, these compounds suffer from several drawbacks. Small molecule drug discovery programs have revealed several allosteric modulators of GLP-1R, such as the quinoxaline “compound 2” (C2) and BETP, which may present alternative approaches for β cell targeting. C2 has been shown to enhance the affinity of several GLP-1R orthosteric ligands and could serve as a “tracer uptake enhancer” for GLP-1R radioligands. Alternatively, radiolabeled BETP derivatives may be of interest as GLP-1R targeting radiotracers, as BETP has been shown to covalently bind to an allosteric site on GLP-1R. This would slow the clearance of the tracer from GLP-1R expressing tissue and may ultimately enhance the target-to-background ratio of the resulting PET image. The investigation of these alternative approaches to β cell imaging represents one of the main aims of this thesis.
The automated radiosynthesis of any novel imaging probe relies on the availability of a well-established and optimized radiochemical toolbox. Many novel probes, such as BETP, present no obvious site for 18F radiolabeling using traditional late-stage radiochemical methodologies; however, the recent disclosure of copper-mediated radiofluorination (CMRF) chemistry has allowed unprecedented access to 18F-labeled electron-rich and -neutral aromatic compounds. These multicomponent reactions are, however, complex and depend on several interacting experimental factors for success. We thus investigated several variations of the CMRF reaction using “design of experiments” (DoE), a statistical approach to reaction optimization, to glean new insights into the behavior of CMRF chemistry. This work reports on the successful optimization of CMRF chemistry for the automated synthesis of [18F]F-BETP and several other novel tracers.