Towards KRAS-directed therapy : Dependency of metastatic colorectal cancer cells on mutant KRAS

The aims of this thesis were 1) to assess the dependency of late-stage colorectal carcinoma (CRC) cells on mutant KRAS and 2) to test the potential of reovirus T3D and COX-2 inhibitors as RAS-targeted therapeutics in experimental models of CRC and colorectal liver metastases. The role of oncogenic RAS in the formation of colorectal liver metastases is evaluated in chapter 2 . We give an overview of the existing literature of both experimental (in vitro and in vivo) and clinical studies that address this issue. Imaging of tumor growth, tumor characteristics and the effect of therapeutic interventions in living animals (intravital imaging) has been made possible by new molecular and optical techniques. In chapter 3 we validate bioluminescence imaging as a minimally invasive tool to monitor tumor growth in the liver. KRAS is important as an initiator of CRC tumorigenesis. In chapters 4 and 5 we asses the dependency of late stage CRC cells on mutant KRAS. We analyze this with a highly aggressive CRC cell line that harbors an endogenous KRAS mutation. In chapter 4 we assess the effect of endogenous mutant KRAS on the interplay between CRC cells and the immune system. In chapter 5 we analyze which distinct stages in the process of liver colonization are affected by endogenous mutant KRAS. RAS proteins signal through a number of distinct signaling cascades. COX-2 is important in the development of multiple cancers, including CRC and bladder carcinoma. The exact relationship between endogenous mutant RAS and COX-2 is unclear. In chapter 6 we investigate this relationship in CRC cells. Furthermore, we investigate the potential therapeutic effects of selective COX-2 inhibitors in a mouse model of established CRC liver metastasis in chapter 6 and in a mouse model of bladder carcinomas in chapter 7. In the ensuing chapters we focus on reovirus, one of the most promising RAS-directed therapeutics. We investigate the mechanism underlying the RAS-specificity of tumor cell killing in chapters 8 and 9 . Next, in chapter 10 we assess the potential of reovirus as a therapeutic agent against experimental CRC liver metastases and investigate the role of the immune system on the therapeutic effect. In chapter 11 we investigate the susceptibility of freshly resected human liver metastases to reovirus T3D infection. We conclude that late stage CRC cells are dependent on mutant KRAS for their invasiveness, metastatic potential and for immune evasion. This makes mutant KRAS an interesting therapeutic target with great potential against metastatic colorectal carcinomas. COX-2 is a direct target of mutant KRAS and selective COX-2 inhibitors have therapeutic potential against colorectal liver metastases and bladder carcinomas. Mutant KRAS sensitizes CRC cells to reovirus T3D induced apoptosis. The therapeutic efficacy of reovirus T3D against established liver metastases is hampered by the host immune system but can be increased by concomitant immunosuppression. The mislocalization of the reovirus receptor JAM-1 in human colorectal liver metastases may be an obstacle for its use as a oncolytic agent in the clinic.