Cancer diagnosis hinges on the accurate early identification of masses and metastases. Treatment likewise demands accurate assessment of the location of any growths and of their size. Such requirements are intimately tied to medical imaging technology, among which molecular tracer-based methods such as Positron Emission Tomography (PET) and Hyperpolarized Magnetic Resonance Imaging (MRI) have emerged as major modalities. These methods rely on the detection of short-lived, isotopically labeled compounds (known as tracers), which serve as reporters on the spatial distribution of their associated biological pathways.

The dominant workhorse of PET is [18F]-Fluorodeoxyglucose (FDG), a tracer whose homology with native glucose allows the imaging of glucose transport. This compound, along with its associated PET-CT imaging technology, has revolutionized our ability to detect and monitor the progression of cancer. Despite this progress, FDG remains limited by high background uptake (especially in inflammatory lesions), which masks authentic tumoral accumulation and decreases confidence in the interpretation of the results. Identification of replacements for FDG have largely focused on other 18F compounds, but few candidates have been advanced to the clinic. This should be unsurprising to some extent because C-F linkages are absent in native human metabolism, leading to changes in behavior despite the fact that fluorine is considered “minimally perturbative.”

These considerations lead naturally to the implementation of carbon-11 PET tracers and carbon-13 MRI tracers, because the unperturbed parent metabolite can be directly prepared. However, synthetic limitations continue to hinder the preparation of novel carbon tracers. The [11C] or [13C] isotope is almost exclusively incorporated via nucleophilic substitution from iodomethane as an N-, S-, or O-methyl functional handle. Though effective, this limits the classes of tracers one can prepare, and moreover presents the radioisotope as a metabolically labile appendage, with in vivo oxidative processing affording the demethylated, “cold” surrogate often on timescales competitive with imaging.10

My laboratory has been focused on solving some of the synthetic limitations associated with carbon-11 and carbon-13 by developing new reagents for the radiosynthesis of arenes with isotopic labels in “core” positions.

Ongoing Experiments:

Our work in the past year has been quite fruitful. At the close of the last reporting period, we had started a new approach, modifying the originally envisioned metallabenzene reagents and instead performing experiments with the closely related metallacyclohexadiene reagents. Though this required a switch from carbon monoxide to carbon dioxide as the isotopic source, this turned out to be beneficial, as for carbon-11 this is actually the proximally produced product of cyclotron bombardment.

In making this change, we developed a synthesis of a new 1,4-pentadienyl-1,5-dilithiate reagent that we used to prepare a series of metallacyclohexadienes. We decided to test the reactivity of these unusual dilithiates directly and were surprised to find that they were in fact more reactive than the transition metal complexes we had been focusing on originally. In reactions with dilute carbon dioxide gas streams, we were able to prepare the corresponding phenols in high yields from a number of substituted precursors.

In parallel, we developed an interest in the related carbon-13 labeled compounds. Since these isotopes are indefinitely stable and can be handled on the benchtop with routine synthetic

procedures, we sought a more convenient one-carbon precursor that maintained the oxidation state of CO2. We subsequently have found that carbon-13 labeled potassium carbonate (commercially available) can be converted in high yield in a single step to the corresponding carbon-13 labeled dibenzylcarbonate. Subsequent reaction of this reagent with our dilithiate reagents generates the corresponding phenols in high yield and in exceptional isotopic purity, with exclusive labeling at the ipso carbon. We have completed a small scope and are preparing an initial publication on this work.

We are also continuing our collaboration with Rich Friefelder at the University of Chicago Cyclotron Facility to optimize the corresponding carbon-11 protocol for phenol synthesis. This work will likely be the subject of a separate publication.

Expansion Plans:

Because of the potential use of carbon-13 labeled arenes as tracers for hyperpolarized MRI imaging, we have initiated a collaboration with Dr. Robert Flavell at UCSF to begin preliminary in vivo experiments with our novel tracer compounds.

We are also actively exploring the application of our dilithiate reagents to carbon-14 labeled phenols (a beta emitting isotope of use in ADME studies). The challenge in this space is the activation of isotopically labeled barium carbonate precursor in an analogous manner – its exceptional insolubility is the primary hurdle to enable productive reactivity.