Activating mutations in the KRAS proto-oncogene occur in 30%-50% of lung adenocarcinomas, the most common subtype of non-small cell lung cancer (NSCLC). Kras is a membrane-associated GTPase that activates multiple kinase pathways, several of which have transforming activity in cellular models. Which of these downstream mediators of Kras contribute to lung tumorigenesis has not been fully elucidated. Our global hypothesis is that KRAS mutations in alveolar epithelial cells promote lung tumorigenesis through two separate mechanisms. First, KRAS mutations transform alveolar epithelial cells. Second, KRAS mutations promote the formation of a tumor microenvironment by increasing the secretion of chemokines that recruit inflammatory cells and vascular endothelial cells to the alveolar epithelium. With respect to the first mechanism, we propose to test the hypothesis that oncogenic KRAS increases the expression of epiregulin, activating ErbB3/EGFR heterodimeric complexes, which transforms the alveolar epithelium. We will test this hypothesis in the first Specific Aim, which is to determine whether loss of epiregulin or ErbB3 is sufficient to block malignant progression in mouse models of KRAS-induced lung cancer, which we will achieve by mating these mice with other mice that are deficient in epiregulin or ErbB3. With respect to the second mechanism, we propose to test the hypothesis that oncogenic KRAS increases the secretion of CXCR2 ligands, recruiting inflammatory cells and vascular endothelial cells to premalignant lung lesions, which is required for lung tumorigenesis. We will test this hypothesis in the second Specific Aim, which is to determine whether CXCR2 loss is sufficient to block the recruitment of inflammatory cells and vascular endothelial cells to premalignant lung lesions and to inhibit malignant progression in mouse models of KRAS-induced lung cancer, which we will achieve by mating these mice with other mice that are deficient in CXCR2. The relevance of these mouse models to human NSCLC will be investigated by analysis of a tissue microarray of 400 NSCLC specimens. We hope to improve treatment for patients with NSCLC, for which there is currently no effective approach, by showing that CXCR2 and its ligands are potentially effective targets in tumors with KRAS mutations and that high intra-tumoral expression of ErbB3 and epiregulin predict efficacy in patients treated with EGFR inhibitors (gefitinib or erlotinib).
