Lung cancer is the third most common cancer in the United States and is the leading cause of cancer-related mortality for both men and women, with 171,500 new cases and 165,600 deaths projected for 1998 in this country. Although smoking cessation trends among adults may result in fewer lung cancer cases in the short to intermediate term, smoking rates among young adults seem to be on the rise. In addition, smoking rates in other parts of the world guarantee that lung cancer will remain a major public health problem worldwide for years to come. One reason for the poor rate of survival of individuals diagnosed with lung cancer is the frequently advanced stage of the disease upon presentation. Thus, there is a clear need to characterize markers associated with early-stage lung cancer to improve methods of diagnosis. Novel therapies for the treatment and prevention of the disease are also wanting. The mouse has been used to model human lung cancer extensively over the past several years in a variety of circumstances, inducting in spontaneous tumor models, chemically-induced models using a variety of agents, and in certain transgenic strains. Although these studies have confirmed that the mouse can be an adequate model of the early stages of human non-small cell lung cancer (NSCLC) (the predominant form of human lung cancer), the methods used in the derivation of these models (e.g., treatment with chemical carcinogens or transgenic expression of viral oncoproteins) and their limited stage of progression has lessened enthusiasm somewhat. This MMHCC team proposes to create novel mouse models of human NSCLC and to characterize these strains in a variety of ways directed toward their proper validation. The strains will harbor targeted activating mutations in the K-ras oncogene (frequently mutated in human NSCLC) and constitutive or conditional loss-of-function mutations in the lung tumor suppressor genes p53 and the Ink4A locus. Existing and future mouse models of NSCLC will be examined by histological analysis in consultation with a veterinary and human clinical pathologist. Mouse tumors will be subjected to extensive DNA and RNA analysis, in part with reference to known molecular changes in human NSCLC and in part in an effort to identify changes that can then be screened for in human tumor samples. The mouse models of NSCLC will be used in cigarette smoke exposure studies and will be used to evaluate investigative chemotherapeutic and chemopreventative agents. A genetic modifier screen will be performed as part of the characterization of one of these strains. Finally, these mouse lung cancer models will be used in the development of novel non-invasive imaging protocols for primary lung tumors and metastatic lesions.