Preclinical Evaluation of a Novel PET Tracer for Imaging PARP1 in Glioblastoma

152 Introduction: Glioblastoma tumor cells can utilize DNA repairing process mediated by poly(ADP-ribose) polymerase-1/2 (PARP-1/2). Thus, partly due to the increased activity of PARPs in tumors, inhibition of PARP-1/2 can potentiate the treatment effects of chemotherapy and radiation therapy.(1,2) Non-invasive quantification of baseline PARP expression may provide prognostic information and guide more precise treatment. Herein, we report the preclinical evaluation of [11C]N-Me-veliparib (NMV) as a PET radiotracer to image PARP-1 in the syngeneic RG2 rat glioblastoma model. Methods: Fischer 344 rats of mixed sex were orthotopically implanted with RG2 cells for 2-3 weeks (n = 14). Dynamic scans of 60 or 90 min duration were performed on the FOCUS-220 scanner in nine rats (baseline, n = 5; blocking with 5 mg/kg i.v. veliparib, n = 4) after [11C]NMV injection. Ex vivo biodistribution studies were performed in seven rats (baseline n = 4, blocking with 5 mg/kg i.v. veliparib, n = 3) at 60 min post-radiotracer injection. Further, the dissected tissues were preserved for immunohistochemistry staining, and protein was extracted for western blotting. Finally, a two-hour dynamic scan was performed on a Focus-220 scanner in a rhesus monkey to evaluate brain uptake of the radiotracer under baseline condition. Arterial blood samples were collected at 5, 15, 60 min after radiotracer injection for metabolite analysis. Results: The radiotracer [11C]NMV was obtained using [11C]MeI in 45 ± 11% decay-corrected radiochemical yield (n = 7), with high radiochemical purity (>97%) and molar activity (141.44 ± 87 MBq/nmol). [11C]NMV showed tumor SUV(40-60 min) of 1.0 ± 0.2 (n = 5) in the baseline scans, compared to tumor SUV of 0.48 ± 0.09 in the blocking scans (n = 4, p = 0.0046) (Fig.1). Biodistribution studies provided uptake in known PARP1-positive tissues, such as tumor (0.85±0.3 %ID/g, n= 4) and spleen (2.85±0.3 %ID/g, n = 4), which were reduced significantly by 75% (n = 3, p=0.020) and 74% (n = 3, p=0.0097) with pre-administration of veliparib, indicating PARP1 specific binding of the radiotracer. Tumor samples from the RG2 rats showed higher expression of PARP1 in western blotting analysis in comparison with other non-tumorous brain regions, supporting the PET imaging results. PARP1 IHC of sagittal plane of whole brain showed overall darker staining due to highly clustered fast proliferating tumor cells with high PARP1 expression. Metabolism of the radiotracer was observed in the metabolite analysis study, with 65% parent tracer remaining in the plasma at 60 min. The major radiometabolite was likely to be the oxidized derivative of the radiotracer, based on preliminary HPLC analysis. The radiotracer showed a relatively low uptake (SUV ~1) in nonhuman primate brain, consistent with the low expression level of PARP1 in monkey cerebral cortex (3). Conclusions: In vivo imaging and ex vivo biodistribution results demonstrated PARP1-specific binding of the novel radiotracer [11C]NMV and its ability to detect PARP1 positive tumor in brain. Western blotting and IHC experiments in vitro further corroborated the imaging and biodistribution findings. [11C]NMV is, to the best of our knowledge, the first brain penetrant PARP PET tracer. References1. Bindra RS. Penetrating the brain tumor space with DNA damage response inhibitors. Neuro Oncol. 2020;22:1718-1720. 2. Alvarado-Cruz I, Mahmoud M, Khan M, et al. Differential immunomodulatory effect of PARP inhibition in BRCA1 deficient and competent tumor cells. Biochem Pharmacol. 2020;184:114359. 3. Ferreira MT, Berger L, Rouleau M, Poirier GG. Assessment of PARP-1 Distribution in Tissues of Cynomolgus Monkeys. Journal of Histochemistry & Cytochemistry. 2020;68:413-435.