The targeted inactivation of oncogene can elicit robust cell death suggesting that tumors can be "addicted" to a mutated proto-oncogene. Understanding why and predicting when tumors are addicted to oncogenes would have protean implications for the development of therapies for cancer. This would enable a more rational basis for choosing molecular targets, identify potentially new non-oncogene targets, facilitate thescreening for potential agents and provide a strategy for selecting patients most likely to benefitfrom specific targeted therapeutic. To study oncogene addiction, we have used the Tet system to develop conditional transgenic mouse models. We have created several conditional oncogenes (MYC, RAS, BCL-2, BCR-ABL), to create transgenic mouse models of different cancers including: T-acute lymphoblastic leukemia (T-ALL), acute myeloid leukemia (AML), osteogenic sarcoma (OS), and hepatocellular carcinoma (HCC). We have found that the consequences of oncogene inactivation are dependent upon both cellular and genetic context. Oncogene addiction involves tumor cell intrinsic and host dependent programs including the permanent loss of self- renewal or induction of cellular senescence and host-dependent programs. We are in the position to now define more generally how these mechanisms contribute oncogene inactivation induced cel death. Our hypothesis is that oncogene addiction can be modeled as a differential response between cel death and survival signaling. We will define key lynch pin gene products and pathways. Our approach will be to perform a quantitative in situ analysis using intravital microscopy and imunohistochemistry combined with a comparative proteomic and genomic analysis. We will perform these studies using different conditional oncogenes and types of cancer in different genetic contexts and then use mathematical modeling and computational biological approaches to reveal the common lynch pin genes and define their mechanistic role in oncogene addiction.