Lung cancer is the foremost cause of cancer related deaths in the U.S. It is a heterogeneous disease composed of genetically and phenotypically distinct tumor cells surrounded by heterotypic cells and extracellular matrix dynamically interacting with the tumor cells. Research in lung cancer is often restricted to patient-derived tumor specimens, in vitro cell cultures and limited animal models, which fail to capture the cellular or microenvironment heterogeneity of the tumor. Therefore, our knowledge is primarily focused on cancer-cell autonomous aberrations. For a fundamental understanding of lung cancer progression and an exploration of therapeutic options, we focused our efforts to develop an Ex Vivo Tumor platform to culture tumors in 3D matrices, which retains tumor cell heterogeneity arising due to in vivo selection pressure and environmental influences and recapitulate responses of tumor cells to external manipulations. To establish this model, implanted syngeneic murine tumors from a mutant KRAS/p53 model were harvested to yield multicellular tumor aggregates followed by culture in 3D extracellular matrices. Using this system, we identified Src signaling as an important driver of invasion and metastasis in lung cancer and demonstrate that EVTs are a robust experimental tool bridging the gap between conventional in vitro and in vivo models.