Abstract IA5: Developing therapeutics in mice with ductal pancreatic cancer

Abstract

Abstract Pancreatic ductal adenocarcinoma (PDAC) is the most lethal common malignancy, with little improvement in patient outcomes over the past 40 years. PDAC frequently harbors somatic mutations in 4 genes (KRAS, P16, Tp53, and SMAD4), and recent exomic and whole genome sequencing efforts have identified a number of less frequent events. Accurate mouse models of PDAC were generated over the last decade, confirming the relevance of these 4 genes and enabling the investigation of fundamental aspects of PDAC tumor biology and therapeutic response. The poor response of pancreatic cancer patients to systemic agents is not predicted by xenograft and other transplanted model systems. An evaluation of this discordant behavior revealed that transplanted tumor models have superior tissue perfusion and delivery of the chemotherapeutic gemcitabine compared to primary murine PDAC, and that this inversely correlates with stromal content. Human PDAC was subsequently confirmed to phenocopy the mouse model as it is also profoundly hypovascular. Stromal depletion with a hedgehog pathway inhibitor increased vascular density and gemcitabine delivery, and prolonged survival to demonstrate a causative role of the hypovascular state in drug responsiveness in PDAC3. The clinical translation of this observation has been encouraging, with several dramatic responses in patients with metastatic PDAC. A randomized trial is currently underway. Alternative methods that target the tumor stroma may also be beneficial in PDAC, including the use of nab-paclitaxel (Abraxane), an albumin-paclitaxel formulation proposed to be sequestered in the intratumoral stroma. Similar to the hedgehog inhibitor, nab-paclitaxel also elevated the gemcitabine levels in the mouse PDAC tumors. However, the mechanism employed was distinct as nab-Paclitaxel did not deplete the stroma and increase the vascular density. Rather, paclitaxel elicits the liberation of reactive oxygen species (ROS), and ROS destabilize and induce the destruction of cytidine deaminase, the major pathway of gemcitabine inactivation in cells4. This novel method of drug synergy was overlooked in the PDX (patient derived xenograft) models, as the chimeric murine stroma is only loosely associated with the human neoplastic cells and they therefore may provide misleading results5. Nab-Paclitaxel treatment with gemcitabine is active in early clinical trials, and genetic variants affecting gemcitabine metabolism may modify this outcome. PDAC is notoriously resistant to VEGF targeted therapeutics despite numerous preclinical experiments that predicted efficacy, and recent approaches in murine PDAC confirmed the prior clinical data6. The hypovascular content of PDAC suggests that the intratumoral endothelial cells are poorly responsive to the available VEGF ligand, and therefore additional pathways may play a more dominant role in vessel proliferation and maintenance. The notch pathway has previously been implicated in vessel morphogenesis, and can be chemically inhibited with gamma secretase inhibitors (GSI). GSI treatment induced the rapid destabilization of the PDAC vasculature, by promoting the death of intratumoral endothelial cells. These effects were exacerbated by concomitant exposure to gemcitabine, resulting in hypoxic necrosis and death of both endothelial cells and neoplastic cells7. A clinical trial has recently begun to evaluate this approach. Although the unique attributes of the PDAC microenvironment participate in the resiliency of PDAC to therapeutics, they also serve as vulnerabilities to exploit for clinical benefit.