Treatment outcomes and prognostic accuracy for oral tongue cancer has had little substantial improvement in the last 20 years. Adjuvant therapy toxicity is high, and there are no clinical tests that accurately stratify which high-risk cancers are likely to benefit, and which will recur despite treatment. Genomic instability is an inherent force in all cancer, however there are substantial differences in which defective repair pathways cause malignancy and contribute to invasion and metastasis.Tongue cancer is relatively aetiologically homogenous, as it is associated mostly with tobacco exposure, exhibits early lymphatic invasion, and is sensitive to particular DNA-damaging agents, making it an attractive disease target for investigating the prognostic potential of genomic stability markers. The aim of this study is to evaluate the prognostic and predictive implications of genomic instability in primary oral tongue cancer.Molecular tumour diagnostics is a burgeoning field, complicated by a paucity of wellpowered trials demonstrating clinical outcomes. DNA repair proteins as biomarkers have been mostly studied in genetic assays, from severe repair syndrome phenotypes to BRCA gene inactivation in breast and ovarian cancer. De-novo mutations in repair genes have been less frequently characterized across sporadic or exposure-related cancers. Almost all cancers exhibit mutation of nuclear DNA resulting in deregulation of the cell cycle and repair process. The most deleterious lesion is the double-strand break (DSB), and defects in this repair pathway cause both increased aggressive features (heterogeneity, invasion and metastasis) and sensitivity to chemoradiotherapy. This study will examine eight markers of genomic stability across three principal domains;1. Markers of repair of specific lesions induced by standard chemotherapy for oral tongue cancer (treatment prediction) 2. Markers implicated in genetic susceptibility studies in oral squamous cell cancer (phenotype stratification) 3 3. Markers of the DNA damage response and DSB repair (genomic stability)A key study strategy is the creation and utilisation of a tissue microarray resource to economically assess the relatively large number of markers in the context of limited source tissue. A microarray of fifty-five (55) included patients was constructed from stored surgical specimen blocks. Immunohistochemical stains were performed, and grading scales constructed under the guidance of a clinical pathologist (Dr Duncan Lambie).The hypothesis of this study is that genomic instability of primary tumours is a characteristic that is important for both prognosis and treatment prediction, and that this can be quantified using immunohistochemical methods. We have identified two important patient subgroups that show significant prognostic differentiation in our markers and discuss the implications for clinical management. I acknowledge that an electronic copy of my thesis must be lodged with the University Library and, subject to the policy and procedures of The Univ...