Repurposing HDAC and mTOR inhibitors for pancreatic cancer.

Abstract

e16234 Background: Pancreatic ductal adenocarcinoma (PDAC) will be the second leading cause of cancer death by 2030 through early onset dedifferentiation and metastasis that results in limited treatments. FDA-approved targeted therapies–including proteasome, mitogen-activated protein kinase (MEK), histone deacetylase (HDAC), and mammalian target of rapamycin (mTOR) inhibitors–have shown preclinical benefit but failed in clinic. However, combination targeted therapy has been largely unexplored in clinical trials. We hypothesized based on prior drug screening efforts in the lab on aggressive subtypes of pancreatic cancer that combinations of these targeted therapies could be synergistically active in preclinical PDAC models. Methods: We optimized combinations of proteasome, MEK, mTOR, and HDAC inhibitors by performing cell viability experiments in human PDAC lines in 2D and 3D cell culture conditions in vitro. Optimized drug combinations from these experiments were then trialed in two different in vivo animal models–human FG xenografts in immunocompromised mice (NSG) and spontaneous primary PDAC tumors from transgenic pancreatic-specific Kras-overexpressing, P53-deficient mice (KPC). RNA-sequencing and Seahorse metabolic assays were utilized to analyze molecular mechanisms of drug synergy. Dedifferentiation was assessed by histology of KPC tumors. Results: The HDAC inhibitor Panobinostat and mTOR inhibitor Everolimus synergistically killed human pancreatic cancer cell lines grown in 2D and 3D cell culture in vitro. Synergy was highest at low concentrations of Panobinostat. In vivo, low-dose Panobinostat and Everolimus synergistically blocked growth of human FG xenografts in NSG mice. In vivo synergy of low-dose Panobinostat and Everolimus was highest in the KPC mouse model, where only combination therapy could significantly reduce tumor growth, stall dedifferentiation, and improve survival. Mechanistically, RNA-sequencing of Panobinostat/Everolimus-treated FG cells revealed that Panobinostat increased tumor suppressor genes opposing dedifferentiation such as p21, EGR1, and CCN2. Everolimus complemented Panobinostat by decreasing expression of enzymes metabolizing acetyl groups such as fatty acid synthase, stearoyl-CoA desaturase, and sterol response element binding factors. Functionally, Everolimus’ activity to reduce oxidative consumption increased histone acetylation specifically in the presence of Panobinostat. Conclusions: A combination of FDA-approved, targeted oral therapies (low-dose Panobinostat/Everolimus) is synergistically active in preclinical models of pancreatic cancer in vitro and in vivo. mTOR inhibitors create drug synergy by specifically increasing histone acetylation in HDAC inhibitor-treated cells by reversing HDAC inhibitor de-repression of acetyl/oxidative metabolism, illustrating a novel connection between cancer epigenetics and metabolism.