“Self-blocking” for PET imaging of cellular proliferation in the liver

Objectives: [¹⁸F]fluoro-3’-deoxy-3’-D-fluorothymidine (FLT) is a PET tracer targeting cytosolic thymidine kinase (TK1) for imaging cellular proliferation. Similarly, 2-chloro-9-(2’-deoxy-2’-[¹⁸F]fluoro-β-d-arabinofuranosyl)adenine ([¹⁸F]clofarabine, or CFA) was developed targeting deoxycytidine kinase (dCK), and 2’-[¹⁸F]fluoro-5-methyl-1-β-D-arabinofuranosyluracil (FMAU) for targeting mitochondrial thymidine kinase (TK2). However, rapid hepatic tracer accumulation resulted in a high liver background signal, which renders these radiolabeled nucleoside analogs not suitable for imaging liver cancers. The mechanism for this high liver background tracer uptake is multi-faceted, not only due merely to hepatic glucuronidation of these nucleoside analogs. To reduce the high liver background activity, “self-blocking” was investigated by using the corresponding unlabeled “cold” compound before radiotracer injection. Using a large amount of unlabeled compound to pre-saturate possible metabolic pathways is not a new idea, and has been studied for a number of tracers under the assumption that the targets in the organ/tissue of interest (e.g., tumor) is more active than in the surrounding tissues, and would elude saturation to certain degree; consequently, “self-blocking” would reduce the background activity. We investigated this with FLT, CFA and FAMU for imaging of liver tumors.

Methods: The human-equivalent dose of unlabeled FLT, CFA and FAMU was applied to a clinically-relevant median-sized animal model of liver cancer in the woodchuck (Marmota monax) 5-10 minutes before the administration of the corresponding radiotracers to examine the changing biodistribution in the liver via PET imaging. [¹⁸F]FLT was synthesized automatically with the Scintomics radiochemistry module (Fürstenfeldbruck, Germany) using the vendor provided protocol. [¹⁸F]CFA and [¹⁸F]FMAU were synthesized automatically using the ELIXYS module (SOFIE, Culver City, CA) with satisfactory radiochemical yield and purity. A clinical Ingenuity PET/CT scanner (Philips, Cleveland, OH) was used for imaging the spontaneously developed HCC in woodchucks. For each scan, list-mode acquisition started upon bolus injection of a radiotracer and lasted for 60 minutes. The dynamically acquired data were binned into 10 X 0.5, 5 X 1 and 10 X 5 min frames and reconstructed with CT-based attenuation correction.

Results: Baseline PET confirmed increased liver background activity of these radiotracers in the woodchuck model. In comparison, pharmacological doses of unlabeled FLT expedited the overall clearance of [¹⁸F]FLT from the liver (< 10 min) and resulted in lower kidney and gallbladder activity as well. Cold CFA administration before [¹⁸F]CFA injection reduced the liver background uptake by 34-47%. The tumor uptake did not significantly change in both cases compared to baseline. In contrast, “self-blocking” seemed to be less effective for FMAU

Conclusions: Application of unlabeled FLT reduced the [¹⁸F]FLT uptake throughout the body. However, therapeutical dose of FLT has been applied to humans with a questionable safety profile. Titration and timing for the application of cold FLT needs to be investigated further for clinical application of “self-blocking” using large dose of cold FLT. Unlabeled CFA did reduce the liver background activity of [¹⁸F]-CFA as expected. CFA was conditionally approved to treat pediatric acute lymphoblastic leukemia, and can potentially be combined with [¹⁸F]CFA for clinical PET imaging of dCK-dependent proliferation in liver cancers. Based on the fact that about half of the TK2 actively exists in the cytosol, not inside the mitochondria, additional study is warranted to develop strategies for the reduction of liver background uptake of FMAU.