Abstract:
Cancer is a life-threatening disease. Its treatment is challenging due to the disease’s heterogenous nature, and there is no universal therapy. In the last decades, different features of cancer have been explored for individualized treatment. One of those attributes is elevated levels of cellular stress, in particular replicative stress, within the tumors. Although treatments targeting replicative stress are available, to date, there is no specific biomarker available for the use with standard-of-care non-invasive imaging methods like position emission tomography utilizing radiolabeled molecules. Thus, therapy approaches are based on ex vivo information like histopathology or are applied without functional guidance.
In this work, we evaluated PARP enzymes for their potential as biomarker for replicative stress. PARPs are heavily involved in the repair of single-strand DNA breaks, and their inhibition leads to synthetic lethality in tumor entities that lack alternative repair mechanisms. First, we synthesized five different PARP radiotracers, small molecules radiolabeled with the β+-emitting isotope 18F, for comparison of their biodistribution in the same mouse model and to determine the best application. We synthesized [18F]FPyPARP, a logD-optimized variant of [18F]PARPi, to shift the clearance route towards renal excretion, as high liver uptake hampers [18F]PARPi application for liver imaging. Compared to the gold-standards [18F]PARPi and [18F]FTT, [18F]FPyPARP presented with improved liver clearance, and might be an alternative to [18F]PARPi for liver imaging. [18F]Olaparib, an isotopologue of the first approved PARP inhibitor olaparib, was synthesized for direct comparison with the 100-fold more effective second-generation isotopologue [18F]talazoparib. The difference in efficacy here is attributed to the improved trapping capacity of PARP on the damaged DNA that prevents replication restart. In a xenograft model, [18F]olaparib and [18F]talazoparib showed similar biodistribution and PARP targeting, suggesting that the PARP trapping capacity does not influence radiotracer performance. In the overall comparison, target engagement was comparable but the radiotracers differed in non-target tissues; Thus, the choice of radiotracer is solely dependent on the envisioned application.
To evaluate PARP as a biomarker for replicative stress, four in vitro models were probed for correlation of PARP radiotracer uptake with levels of stress. In myc overexpression models, the results were heterogeneous, and another attempt for a mIDH expression-based cell model did not indicate increased uptake. We then set out to induce replicative stress chemically, and did not observe significant differences in PARP radiotracer uptake compared to controls. We concluded that PARP is not a suitable biomarker as the expression is not upregulated but more likely the protein is activated on an enzymatic level upon replicative stress. Several other potential biomarkers were tested for changes in expression levels with Western blot, but did not result in a clear specific biomarker.
As a surrogate, we developed novel reporter gene systems to compensate the lack of a specific replicative stress biomarker for preclinical therapy development and research on biomarkers and animal models. A reporter gene could be used to quantify promotor activity or other biological processes that can not be visualized directly. We designed, characterized and evaluated HaloTag, SNAPTag and CLIPTag and novel radiotracers designed to target the respective proteins in vitro and in vivo in a pilot xenograft study. All three presented with excellent target engagement and favorable pharmacokinetics.
Interestingly, the HaloTag and CLIPTag radiotracers showed unspecific uptake in the naïve rodent brain, indicating that they are able to cross the intact blood-brain barrier. The blood-brain barrier is a recurrent obstacle in brain radiotracer development and thereby hampers global visualization of biological processes. Further evaluation in a murine model of viral gene transfer to the brain confirmed specific brain uptake and paves the way for future applications in the whole body.
Thereby, our novel reporter gene systems are suitable to be used for future development of replicative stress specific radiotracers. The potential to stratify patients according to levels of replicative stress would ultimately aid selecting appropriate therapy regimen, and pave the way for new treatments targeting replicative stress.
Print-on-Demand