Lung cancer is the most common cause of worldwide cancer deaths. Recent mutation based therapeutic developments for patients with lung cancer have occurred in those with adenocarcinomas, and there is an urgent need for improved management of patients with squamous cell carcinoma (SCC), a subtype accounting for 67,000 new patients per year in the US. Furthermore, SCC is smoking related as SCC patients are mostly current or former smokers and severely stigmatized as having a "self-inflicted disease". Research in SCC is severely underfunded (possibly for this same reason). This proposal leverages a multi-institutional and multidisciplinary approach to validate preexisting gene expression signatures for prognosis of SCC, with the goal to bring these signatures to a more broadly clinical applicable assay (i.e. RT-qPCR). However, we also hope to identify new targets for therapy for patients with SCC. The proposal has 3 aims. In Aim1 we will validate previously defined gene array (mRNA) and miRNA expression signatures for prognosis in early stage SCC. We will use 300 SCC tumors collected by different institutions from clinically annotated cohorts with at least 3 years follow-up, and validate mRNA and miRNA signatures in one standardized laboratory per assay platform. In this aim, we will transform these signatures into RT-qPCR based assays. In Aim 2 we will validate the most robust signatures in 2 independent cooperative group clinical trial cohorts; one from CALGB (N=150 pts.) and one from ACOSOG (N=250 pts.). The endpoint of these studies is prognosis (disease related survival). In Aim 3, we will leverage the biospecimen resources described above to begin validation of genomic variants being identified by NCI's Cancer Genome Atlas Project and assess their prevalence and whether they can supplement the prognostic power of the signatures defined in Aims 1 and 2. We have established a Squamous Lung Cancer Consortium (SLCC) of experienced clinical and biomarker investigators and we will establish a Central SCC database including a virtual SCC tissue bank linked to molecular, pathological, and clinical information.