Abstract IA01: Deconstructing K-RAS oncogenic signaling in lung and pancreatic tumors

Abstract

Abstract The presence of K-RAS oncogenes in human tumors has been known for over 35 years. Indeed, extensive analysis of cancer genomes has revealed that they are present in at least a quarter of all solid tumors. Moreover, there is overwhelming evidence that they represent the initiating event in lung and pancreatic adenocarcinomas, two cancer types with some of the worse prognoses. Recent studies in genetically engineered mouse (GEM) tumor models have shown that inhibition of K-RAS oncogene expression results in rapid tumor regression. Yet in spite of recent progress, K-RAS oncogenes remain for the most part undruggable. Ironically, the main K-RAS downstream effectors are druggable kinases. Selective inhibitors against members of the MAPK and PI3K signaling pathways have been developed over the last two decades. Yet none of them has yielded suitable results in clinical trials to merit FDA approval against K-RAS mutant malignancies. In an effort to identify K-RAS vulnerabilities that may have been missed using biochemical or pharmacologic approaches, we decided a decade ago to deconstruct K-Ras oncogenic signaling in GEM tumor models by interrogating the effect of ablating or inactivating its downstream kinases during tumor initiation. In addition, we examined the consequences of systemically eliminating or inhibiting these targets during adult homeostasis, in order to evaluate the potential toxic effects that may result from tampering with the MAPK and PI3K signaling pathways. These studies revealed three classes of downstream targets: those whose ablation had no consequences on tumor development, those that prevented tumor development but were unacceptably toxic when ablated systemically, and a limited number (c-Raf and Cdk4) that prevented or thwarted tumor development without causing significant toxicities upon systemic elimination. In addition to these downstream targets, we were also able to identify the EGFR as a potential target in pancreatic but not in lung tumors. More recently, we have established a new generation of “therapeutic” GEM tumor models in which we can separate, both spatially and temporally, tumor development from target ablation/inactivation by using two independent recombinase systems. In these models, whereas K-Ras or K-Ras/TP53-driven tumor development is mediated by the Flp(o)/frt system, we use the inducible CreERT2/loxP system for systemic target ablation/inactivation. These “therapeutic” models have allowed us to explore the therapeutic potential of these previously validated targets in tumor-bearing mice. Our results have revealed that c-Raf ablation induces significant levels of tumor regression in K-Ras/TP53-driven lung tumors, a therapeutic effect that is further potentiated by the concomitant expression of a kinase dead isoform of Cdk4. In the corresponding GEM pancreatic tumor model, ablation or inactivation of c-Raf, Cdk4, or EGFR failed to prevent tumor regression, although tampering with Cdk4 or EGFR resulted in significant tumor delays. Importantly, concomitant ablation of c-Raf and EGFR in K-Ras/TP53-driven advanced pancreatic tumors resulted in complete tumor regression in half of the mice. Preliminary comparison of these “responder” tumors with those that did not respond to c-Raf/EGFR ablation is revealing some vulnerabilities that may provide additional targets against these advanced tumors. We hope that these results may serve to pave the way to define more efficacious therapeutic strategies against K-RAS mutant tumors that might eventually be translated to the clinic. Citation Format: Mariano Barbacid. Deconstructing K-RAS oncogenic signaling in lung and pancreatic tumors [abstract]. In: Proceedings of the AACR Special Conference on Targeting RAS-Driven Cancers; 2018 Dec 9-12; San Diego, CA. Philadelphia (PA): AACR; Mol Cancer Res 2020;18(5_Suppl):Abstract nr IA01.