The incredibly short survival of patients diagnosed with pancreatic cancer underscores one of the greatest failures of the medical community: our inability to identify effective treatments for pancreatic ductal adenocarcinoma. Most new compounds in clinical trials fail because of highly toxic side effects in normal tissues and limited efficacy in tumor response. An ideal drug would kill only tumor cells and spare the normal cells. In this regard, poly(ADP-ribose) polymerase (PARP) inhibitors have demonstrated a promising therapeutic potential, particularly in patients with BRCA-deficient tumors. BRCA-deficient cells have an impaired homology-directed repair (HDR) pathway, a mechanism critical for resolving DNA double strand breaks (DSBs). When PARP1 is inhibited, DNA single strand breaks are not repaired and they mature to double strand breaks, which are very toxic for the cell. Hence, inhibition of PARP1 is selectively toxic to cells with an impaired HDR pathway, such as BRCA-deficient tumor cells, but not to normal cells. Because up to 10% of pancreatic cancer patients carry mutations in either BRCA2 or other proteins involved in the HDR pathway, PARP inhibitors may also prove effective in this setting. However, a number of subtleties remain that call this conclusion into question, including poor drug delivery, loss of heterozygosity, and BRCA-independent alteration of the HDR pathway.

We designed a preclinical study to evaluate the efficacy of MK-4827, a potent PARP inhibitor, in three different genetically engineered mouse models representing different pancreatic cancer patient populations. The KPC mice carry mutations in genes that are frequently mutated in pancreatic cancer patients and develop tumors that fully recapitulate the biology of human pancreatic ductal adenocarcinoma. These mice will represent the broad population of patients with spontaneous pancreatic tumors. A second model with heterozygous mutations in BRCA2 (KPCB2+/-) will represent patients with hereditary BRCA2 mutations. A third model with homozygous conditional deletion of BRCA2 (KPCB2f/f) will represent spontaneous tumors that have completely lost HDR functions. We show that BRCA2-deficient KPC tumor cells strongly respond to treatment with MK-4827 in vitro and in vivo. MK-4827 efficiently inhibited PARP activity, increased double strand break formation and induced M-phase cell cycle arrest in tumors of KPCB2f/f mice. In the future, we will use pre- and post-treatment tumor biopsies to determine the molecular mechanisms of response to PARP inhibition. The ultimate goal of this study is to identify genetic determinants of PARP sensitivity and validate our results in a clinical setting. Identification of a biomarker would be an extremely valuable clinical tool to classify pancreatic cancer patients that could respond to the treatment with PARP inhibitors, a tool that is currently not available.

Citation Format: Barbara Orelli, Stephen A. Sastra, Carmine F. Palermo, Thomas Ludwig, Kenneth P. Olive. Preclinical evaluation of a PARP inhibitor in mice representing genetically different subtypes of pancreatic cancers. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 2195. doi:10.1158/1538-7445.AM2013-2195