PET imaging of 11C-labeled cyclooxygenase-1 (COX-1) and COX-2 radioligands in ovarian, breast, and pancreatic cancer xenograft mouse models.

440 Objectives: The role of putative inflammatory biomarkers cyclooxygenase-1 (COX-1) and COX-2 in neuroinflammation as well as cancer progression and prognosis is evident, yet development of PET radiopharmaceuticals to explore these targets has been challenging. [11C]PS13 and [11C]MC1, which target COX-1 and COX-2, respectively, have been evaluated in non-human primates[1], and [11C]PS13 has recently been translated for human use (ClinicalTrials.gov identifier: NCT03324646), to explore neuroinflammation. The aim of this study is to evaluate these two radioligands in xenograft mouse models of ovarian, breast, and pancreatic cancers. Methods: Radiosyntheses of [11C]PS13 and [11C]MC1 were performed as previously described[1] and the tracers were evaluated in a panel of xenograft mouse models prepared from ovarian, breast, and pancreatic cancer cell lines, OVCAR3, MDA-MB-231, and PANC-1, inoculated subcutaneously (s.c.) on the right flank of ICRscid mice. Evaluation of [11C]PS13 and [11C]MC1 was performed by dynamic PET imaging including blocking studies, ex vivo biodistribution, and radiometabolite analysis of plasma and tumor homogenates. Results: OVCAR3 xenografts were well visualized with [11C]PS13 (0-60 min average image). Time-activity curves (TACs) revealed steady tumor radioactivity accumulation that plateaued from 40-60 min with an average uptake of 3.56 ± 0.81 %ID/g (40-60 min), and was significantly reduced by pre-treatment with the known COX-1 inhibitor, ketoprofen, to 1.30 ± 0.18 %ID/g (p=0.0096), and was significantly higher than muscle tissue, 0.99 ± 0.21 %ID/g (p=0.00002). In vivo PET imaging analyses were validated by biodistribution studies. TACs revealed steady tumor radioactivity accumulation with no washout over 60 min. Radiometabolite analysis revealed the parent compound, [11C]PS13, to account for 65% of radioactivity in the tumor, versus 25% in cardiac blood plasma, at 40 min post-injection of the radiotracer. MDA-MB-231 xenografts were not visualized by PET imaging with [11C]PS13 or [11C]MC1 despite reported overexpression of COX-1 and COX-2[2]. PANC-1 xenografts were not visualized by [11C]MC1 despite reported high expression of COX-2[3]. Conclusions: [11C]PS13 has a favorable radiobiological profile for studying the role of peripheral COX-1 and clinical translation of this radiopharmaceutical in oncology is planned. We are concurrently evaluating [11C]MC1 in alternative COX-2 expressing xenograft mouse models. References: [1] Pike, VW & Innis, RB, et al. Evaluation of two potent and selective PET radioligands to image COX-1 and COX-2 in rhesus monkeys. J Nucl Med 2018;59:1907-12. [2] Liu XH and Rose DP. Differential expression and regulation of cyclooxygenase-1 and -2 in two human breast cancer cell lines. Cancer Res 1996;56:5125-7. [3] Li S, et al. Anti-tumor effect and mechanism of cyclooxygenase-2 inhibitor through matrix metalloproteinase 14 pathway in PANC-1 cells. Int J Clin Exp Pathol 2015;8:1737-42.