Identification of Oncogenic KRAS-Associated Vulnerabilities in Non-Small Cell Lung Cancer

Activating mutations in KRAS are frequently involved in the pathogenesis of non-small cell lung cancer (NSCLC), the disease responsible for the most cancer-related deaths in the US. Despite intensive efforts to develop drugs that directly interfere with KRAS activity over the past decade, no effective inhibitor has been developed. As an alternative, synthetic-lethal therapeutic opportunities are being pursued using large-scale, RNAi-based, functional genomics platforms. We first addressed two major challenges associated with RNAi-based primary synthetic-lethal screens; a prevalent miRNA-like behavior of siRNA and cell line-dependent phenotypic diversity within intra-lineage KRAS-driven cancer. In consideration of these, we performed a whole-genome synthetic-lethal siRNA screen, powered by 106 NSCLC lines and integrated with gene set enrichment analysis. This identified components of nuclear transport machinery as selectively essential for KRAS mutant NSCLC lines. We found that pharmacological inhibition of a key nuclear export receptor, XPO1 (a.k.a. CRM1), was sufficient to induce robust and selective apoptosis in KRAS mutant NSCLC cells in vitro and to cause significant impairment of KRAS mutant tumor growth in vivo. Mechanistically, XPO1-depedent nuclear export machinery was required to maintain NFκB-mediated survival signaling. We discovered that a subset of KRAS mutant NSCLC lines bypassed the addiction to XPO1-dependent nuclear export via YAP1 activation as a consequence of previously unappreciated co-occurring loss-of-function mutations in FSTL5 and mutations in Hippo pathway. The intrinsic resistance was reversed by coadministration of YAP1/TEAD inhibitor. Thus, our study suggests that XPO1 can be exploited for a promising therapeutic opportunity for KRAS mutant lung cancer and provides strategies for genomics-guided application of clinically available XPO1 inhibitors.