New Routes to Carbon-11 Based Molecular Imaging Agents for In Vivo PET Imaging

Student thesis: Doctoral ThesisDoctor of Philosophy


Carbon-11 is a short half-life (20.4 min) positron emitting radionuclide which is commonly incorporated into small organic molecules to produce radiotracers for PET imaging. Carbon-11 is generally produced in a cyclotron as carbon dioxide ([11C]CO2), but this is fairly unreactive and so is generally converted to secondary, more-reactive “synthons” such as methyl iodide ([11C]CH3I). Whilst highly useful for incorporating carbon-11 into methyl positions on many molecules, this limits the chemical space that can be radiolabeled with carbon-11 and there is a need for the development of novel 11C- radiochemistry to allow radiolabeling of more diverse molecular structures. Some recently developed methods allow for the direct fixation of [11C]CO2 into organic molecules, to produce carbonyl-11C-radiolabeled compounds. This thesis covers two key implementations of this new chemistry.

The first of these involved the development of a thin-film in-loop setup for [11C]CO2- fixation and 11C-urea synthesis. [11C]CO2-fixation methods involve bubbling [11C]CO2 through fixation solutions, but this step can be difficult to incorporate into a routine automated synthesis due to inefficiencies in [11C]CO2 trapping, as well as losses incurred during transfer of the solution from the vial to an HPLC injector-loop for purification. 
To address these issues, an “in-loop” setup was developed in which [11C]CO2 was efficiently trapped (~ 99%) in a thin-film of amine/DBU solution supported on the inner surface of a short section of polymer tubing. Passing Mitsunobu reagents through this tubing resulted in the rapid and efficient production of symmetric 11C-ureas in-flow. The use of cheap and widely-available components should make the developed methodology straightforward to implement in other labs; while its use of disposable plastic components could make it well suited to GMP production. N,N’-[11C]dibenzylurea (82% RCY), N,N’- [11C]diphenylurea (24% RCY), and N,N’-[11C]dicyclohexylurea (21% RCY) were all produced with this setup, in < 3 min from the EoB. 
The second of these implementations involved the development of a generally-applicable method for 11C-radiolabeling PSMA-targeted radioligands via [11C]CO2-fixation. PSMA is a well-validated molecular target for PET imaging (and increasingly, targeted radioligand therapy) of prostate cancer. There is an ongoing effort worldwide into the development of newer generation PSMA-targeted ligands, to modify their target binding affinities, off-target uptake, and clearance pathways. This effort is hampered by the difficulties in developing custom radiolabeling methods for each new candidate compound. The aim was therefore to develop a method to 11C-radiolabel all of these PSMA ligands within the ubiquitously-present PSMA-targeting glutamate-urea pharmacophore. 

Initially, two different [11C]CO2-fixation and 11C-urea synthesis approaches were explored: DBU/Mitsunobu-mediated, and BEMP/POCl3-mediated. It was found that the (harsher) BEMP/POCl3 method was still compatible with these substrates, whilst being somewhat more efficient and reproducible. Furthermore, during the optimisation of this method, several key factors were considered to ensure the method is useful and widely- applicable: the [11C]CO2-fixation must be efficient, the 11C-urea synthesis must be selective for the asymmetric urea product, and the stereochemistry must be retained. Ultimately, a small diverse library of PSMA-targeted 11C-radioligands were synthesised (13-35% RCYs) in < 14 minutes from the EoB using a single well-optimised and generally-applicable “one-pot” radiosynthetic method. 
Date of Award1 Jun 2022
Original languageEnglish
Awarding Institution
  • King's College London
SupervisorTony Gee (Supervisor)

Cite this