Modern medicine has had a limited impact on pancreatic cancer: drugs targeting signaling pathways are not active, death-rates are increasing, and five-year survival is only 8%. Pancreatic cancer develops through the stepwise acquisition of mutations in KRAS in 95% of tumors, CDKN2A, TP53 and SMAD4. Many of these mutations both drive proliferation and modify cellular metabolism.

Until very recently only three systemic treatments had demonstrated activity in pancreatic cancer: gemcitabine, gemcitabine combined with nanoparticle albumin-bound paclitaxel (Abraxane) and FOLFIRINOX therapy, all of which were used in non-selective patient populations. In 2019 the results of the phase III POLO trial were published (with Dr. Golan as principal investigator) examining the efficacy of the targeted therapeutic drug Olaparib in pancreatic cancer. The trial was unique in several respects: the first time a poly(adenosine diphosphate-ribose) polymerase (PARP) inhibitor (PARPi) has demonstrated activity in a randomized pancreatic cancer - with a doubling of response rate, and the first time biomarkers (germline BRCA1 or BRCA2 mutation) had been used to drive a successful phase III trial with targeted therapy in pancreatic cancer. Among patients with a germline BRCA mutation and metastatic pancreatic cancer, progression-free survival was longer with maintenance Olaparib than with placebo. Intriguingly, a close analysis of the data suggested that there were multiple, discrete groups within the BRCA mutant population with various levels of response to PARPi. The goal of this project is to combine genomic, transcriptomic and metabolomic approaches to identify mechanism of resistance to therapy, both for Cisplatin and PARPi, and identify additional non-BRCA patients that can benefit from treatment with Olaparib.

This project already sequenced RNA from ~300 samples derived from ~100 BRCA1\2 and WT PDX models, as well as performed whole genome sequencing for a couple of dozens models. We have identified different clusters which are significantly associated with lack of response to cisplatin, and computationally characterized their metabolic phenotypes. A few targets have been highlighted, which are now being experimentally validated. Specifically, we previously found that changes in metabolism underlie pancreatic cancer’s resistance to radiation, and we successfully increased cancer cells’ death by modulating their metabolism. We now observed that pancreatic cancer cells that are resistant to radiation display resistance to the chemotherapy drug cisplatin as well, and further examine the effect of specific gene targets emerged from our analysis. In parallel, we work to construct a computationally-derived BRCAness signature which may identify additional patients that can benefit from PARPi treatment.