Genetically engineered mouse models (GEMMs) are powerful pre-clinical tools for translational lung cancer research. These models allow us to address questions that would not be possible using patient samples or cancer cell lines alone. In my presentation I will present our work aimed at gaining insight into what drives intratumoral heterogeneity in lung cancer. In particular, we are interested in how distinct cancer initiating cells and specific genetic alterations can influence the histopathology of lung tumours and their subsequent response to therapies. To achieve this, we have manipulated a number of different oncogenes and tumour suppressor genes (p53, Rb1, KrasG12D, Pten, Keap1, Sox2) in distinct lung epithelial cell types by targeting Cre-recombinase expression specifically to Club cells, neuroendocrine cells, alveolar type II cells and basal cells. Using this technology, we can induce small cell lung cancer (SCLC) and non-small cell lung cancer (NSCLC). We show that multiple cell types can give rise to these tumours, but that the cell-of-origin is an important factor in determining the tumour phenotype. Our data indicate that both cell type-specific features and the nature of the oncogenic lesion(s) are critical factors in determining the tumour initiating capacity of lung epithelial cells. Future work will focus on how we can use this information to devise new strategies for prevention and early detection of lung cancer, as well as identifying pre-malignant clones that may contribute to relapse of this devastating disease.