Abstract A27: Identification of novel KRAS-synthetic lethal targets for treatment of KRAS-dependent Non-Small Cell Lung Cancer

Abstract

Abstract Lung cancer is the number one cause of cancer-related deaths worldwide. The most prevalent type of lung cancer is Non-Small Cell Lung Cancer (NSCLC). A significant number of patients with NSCLC carry oncogenic KRAS mutations. However, the efforts to target KRAS directly have thus far proven unsuccessful and tumors harboring mutations in this gene remain the most difficult to treat, highlighting the need for alternative approaches. One promising avenue to treat these cancers is to exploit the concept of synthetic lethality. To identify novel KRAS synthetic lethal interactions we performed shRNA gene knock-down screens in 3D cultures of cells derived from a mouse model of NSCLC driven by the activation of KRAS and loss of P53. Growing cells in 3D culture was shown to more faithfully recapitulate important aspects of cancer biology when compared to cells grown as monolayers. In addition, 3D cultures were shown to be enriched in tumor-propagating cells (TPCs), a subset of cells driving tumor initiation, maintenance and progression. We have also previously shown that TPCs contribute to chemoresistance and that the TPC-specific gene signature correlates with poor prognosis in NSCLC patients. Here we used pooled shRNA libraries targeting 623 genes associated with tumor response to chemotherapy, DNA damage response, components of the KRAS interaction network, and KRAS effector genes to screen for potential KRAS-dependent TPC vulnerabilities. As a control we performed the screen in KRAS-mutant murine lung cancer cells grown as monolayers to determine which genes are specifically required for survival in 3D culture. These screens identified many potential KRAS synthetic lethal targets, knock-down of which selectively inhibited growth in 3D but had no effect on 2D growth. To validate and characterize several of these genes, we used the CRISPR/Cas9 system in human lung cancer cells. Single-guide RNAs targeting these genes were introduced into Cas9-expressing NSCLC cell lines or immortalized non-tumorigenic human bronchial epithelial cells (HBEC). The cells were subsequently cultured in adherent or 3D conditions to allow for formation of colonies or spheres, respectively. Knock-out of some of these genes selectively impaired 3D growth of KRAS-dependent NSCLC cells, but not of the HBECs and/or KRAS wild-type NSCLC cell lines, confirming their synthetic lethal relationship with KRAS. The most promising of these targets will be further validated in vivo in primary mouse lung tumors. Finally, human relevance will be determined using patient-derived xenograft (PDX) models. This strategy should lead to the development of novel KRAS-centered therapeutics to treat lung cancer and other KRAS-driven cancers. Citation Format: Kaja Kostyrko, David Simpson, Joshua Broyde, Marcus Kelly, Andrea Califano, Barry Honig, Peter Jackson, Alejandro Sweet-Cordero. Identification of novel KRAS-synthetic lethal targets for treatment of KRAS-dependent Non-Small Cell Lung Cancer [abstract]. In: Proceedings of the AACR Precision Medicine Series: Opportunities and Challenges of Exploiting Synthetic Lethality in Cancer; Jan 4-7, 2017; San Diego, CA. Philadelphia (PA): AACR; Mol Cancer Ther 2017;16(10 Suppl):Abstract nr A27.