Initial biological evaluations of 18F-KS1, a novel ascorbate derivative to image oxidative stress in cancer.

Reactive oxygen species (ROS)-induced oxidative stress damages many cellular components such as fatty acids, DNA, and proteins. This damage is implicated in many disease pathologies including cancer and neurodegenerative and cardiovascular diseases. Antioxidants like ascorbate (vitamin C, ascorbic acid) have been shown to protect against the deleterious effects of oxidative stress in patients with cancer. In contrast, other data indicate potential tumor-promoting activity of antioxidants, demonstrating a potential temporal benefit of ROS. However, quantifying real-time tumor ROS is currently not feasible, since there is no way to directly probe global tumor ROS. In order to study this ROS-induced damage and design novel therapeutics to prevent its sequelae, the quantitative nature of positron emission tomography (PET) can be harnessed to measure in vivo concentrations of ROS. Therefore, our goal is to develop a novel translational ascorbate-based probe to image ROS in cancer in vivo using noninvasive PET imaging of tumor tissue. The real-time evaluations of ROS state can prove critical in developing new therapies and stratifying patients to therapies that are affected by tumor ROS.

We designed, synthesized, and characterized a novel ascorbate derivative (E)-5-(2-chloroethylidene)-3-((4-(2-fluoroethoxy)benzyl)oxy)-4-hydroxyfuran-2(5H)-one (KS1). We used KS1 in an in vitro ROS MitoSOX-based assay in two different head and neck squamous cancer cells (HNSCC) that express different ROS levels, with ascorbate as reference standard. We radiolabeled ¹⁸F-KS1 following ¹⁸F-based nucleophilic substitution reactions and determined in vitro reactivity and specificity of ¹⁸F-KS1 in HNSCC and prostate cancer (PCa) cells. MicroPET imaging and standard biodistribution studies of ¹⁸F-KS1 were performed in mice bearing PCa cells. To further demonstrate specificity, we performed microPET blocking experiments using nonradioactive KS1 as a blocker.

KS1 was synthesized and characterized using ¹H NMR spectra. MitoSOX assay demonstrated good correlations between increasing concentrations of KS1 and ascorbate and increased reactivity in SCC-61 cells (with high ROS levels) versus rSCC-61cells (with low ROS levels). ¹⁸F-KS1 was radiolabeled with high radiochemical purity (> 94%) and specific activity (~ 100 GBq/μmol) at end of synthesis (EOS). Cell uptake of ¹⁸F-KS1 was high in both types of cancer cells, and the uptake was significantly blocked by nonradioactive KS1, and the ROS blocker, superoxide dismutase (SOD) demonstrating specificity. Furthermore, ¹⁸F-KS1 uptake was increased in PCa cells under hypoxic conditions, which have been shown to generate high ROS. Initial in vivo tumor uptake studies in PCa tumor-bearing mice demonstrated that ¹⁸F-KS1 specifically bound to tumor, which was significantly blocked (threefold) by pre-injecting unlabeled KS1. Furthermore, biodistribution studies in the same tumor-bearing mice showed high tumor to muscle (target to nontarget) ratios.

This work demonstrates the strong preliminary support of ¹⁸F-KS1, both in vitro and in vivo for imaging ROS in cancer. If successful, this work will provide a new paradigm to directly probe real-time oxidative stress levels in vivo. Our work could enhance precision medicine approaches to treat cancer, as well as neurodegenerative and cardiovascular diseases affected by ROS.