PROJECT SUMMARY Head and neck cancer (HNC) is the sixth most common cancer in the United States. Today we know that this cancer represents two distinct diseases, human papillomavirus positive (HPV+) HNC and HPV negative (HPV-) HNC, with remarkably different molecular profiles and clinical outcomes. Yet, these two diseases are currently treated the same way, using surgery combined with chemoradiotherapy that are commonly associated with significant morbidities and modest overall survival rates for the HPV- cohort. Multiple genes are mutated in HNC. Which of these mutated genes are the true drivers of carcinogenesis in the head and neck, and under what context, is not known. Identifying these genes and exploring biomarkers predictive of disease severity can be achieved using genetically engineered mice. In this project, we will make use of our prior expertise in developing well-validated mouse models for human cancers to identify driver mutations in HNC, and test drugs that target the molecular pathways impacted by these drive mutations. We will use information gained from genome-wide analyses of these mouse models to identify biomarkers that are predictive of disease severity and response to therapy, and test their utility in human HNCs making use of the broad panel of patient derived xenografts (PDXs) recently established at the University of Wisconsin. This project has three specific aims: (1) define mutations that drive head and neck carcinogenesis; (2) identify therapeutic targets for treating HNC; and (3) perform human endpoint analyses using human HNC tissue microarrays to identify biomarkers that are predictive of disease severity and response to therapy. This project is highly innovative and significant because we shall identify mutations that drive HNC, test relevant new targets for therapy and define new prognostic biomarkers for human HNC.