Abstract 4690: A multimodality imaging end-point study of everolimus and ganetespib in treatment of pancreatic cancer: A pre-clinical PET/MRI/MRS study

Background: mTOR is a significant regulator of cell metabolism, proliferation, autophagy, and is strongly linked to cancer cell survival. This suggests that inhibiting mTOR with everolimus (RAD001) may inhibit tumor cell growth while promoting cell death. The aim was to establish imaging end-points for the efficacy of everolimus in combination with the second generation HSP-90 inhibitor ganetespib in everolimus resistant pancreatic cancer (based on promising results from initial patient trials). This study aimed to establish metabolic (FDG-PET and MRS), morphological (DWI), and anatomical (magnetic resonance imaging, MRI) end-points for mTOR and HSP inhibitor response in mouse pancreatic adenocarcinoma xenografts. Methods: Human Panc-159 tumor xenografts were established in athymic nude mice. Everolimus was administered orally at a daily dose of 5 mg/kg; ganetespib was administered via tail vein injection weekly 100 mg/kg. PET/DWI-MRI imaging was done at baseline, mid-study (Day 8-9), and end of study (Day 21-22). At end of study, mice were sacrificed, and tumors were extracted with perchloric acid for use in 1H-MRS metabolomics. Multivariate analysis was conducted with MetaboAnylast2.0 software. Results: Single treatments showed no significant inhibition of tumor growth or decrease in tumor cellularity. Everolimus single treatment did cause a slight decrease (p = 0.0296 by FDG-PET) in glucose uptake- as previously reported for the mTOR pathway inhibition, while ganetespib alone did not alter glucose uptake. Combination treatment inhibited tumor growth (p= 0.0018 by MRI) in addition to decreasing tumor cellularity (p = 0.0323 by DWI). Combination treatment also decreased tumor metabolic activity as demonstrated through the decrease in glucose uptake (p = 0.0085). Multivariate analysis on combined quantitative metabolic and imaging data sets (including tumor volume, FDG-SUV, cellularity, and 1H-MRS derived concentrations of endogenous metabolites) showed group separation between the control and combination treatment groups. Variables with highest importance in projection (VIP) accounting for group separation were tumor cellularity (VIP = 3.0262), tumor volume (VIP =2.3178), phospholipid-associated glycerol (VIP = 1.9073), and FDG-SUV (VIP = 1.3406). Conclusions: Our study provides the rational of combining everolimus with ganetespib to overcome drug resistance and increase treatment responsiveness in pancreatic cancer. Combination treatment did result in decreased tumor growth, cellularity, phospholipid metabolism and glucose uptake giving potential for these variables to serve as effective biomarkers for therapeutic response once validated in clinical trials. Even though Panc-159 is highly resistant to single everolimus treatment, this combination treatment with an HSP-90 inhibitor provides first evidence of proliferation and metabolic response. Citation Format: Justin Y. Lee, Lora A. Wilson, Jerry Choi, Kendra M. Huber, Andrea L. Merz, Colin D. Weekes, Natalie J. Serkova. A multimodality imaging end-point study of everolimus and ganetespib in treatment of pancreatic cancer: A pre-clinical PET/MRI/MRS study. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 4690. doi:10.1158/1538-7445.AM2014-4690