Cancer Control Research5R01CA095419-07
ORAL CANCER: MOLECULAR PROFILES & CLINICAL OUTCOMES
Oral squamous cell carcinoma (OSOC) is associated with considerable mortality and morbidity. The current staging system does not adequately predict the clinical outcome of individual 0SCC patients. In this ARRA application, we propose to test the respective gene expression signatures for the prediction of invasive 0SCC, nodal metastasis, and 0SCC-specific survival that we identified in our initial study (R01CA095419) using banked samples and additional samples from newly enrolled patients. Our ultimate goals are to address three clinically important questions: 1) What determines which 0SCC patients go on to develop local recurrence and/or second primary tumors, so that more aggressive medical treatment and/or surveillance can be offered selectively to these patients? 2) How can we identify those clinically node-negative 0SCC patients who harbor occult nodal metastasis, patients who should undergo neck dissection while those without can be spared unnecessary surgery? 3) To what degree can molecular markers improve upon the prediction of survival based on the current staging system so that physicians can better individualize patients' treatments? In the 2-year ARRA funding period, we plan to continue to enroll 0SCC cases and collect biological samples and comprehensive data to support the following aims: 1a: Validate our discovery of occult metastasis-related genes in another set of clinically node-negative 0SCC patients for their ability to predict the presence of occult metastasis or the development of nodal metastasis. 1b: Validate the predictive ability of the genes we found most strongly related to OSCC-specific survival in a new group of patients, and validate our observation that our genetic expression signature for survival when combined with TNM stage predicts 0SCC-specific survival better than TNM stage alone. 2: Test the surgical margins and clinically normal oral mucosa of 137 0SCC patients by qRT-PCR for the expression of four genes in the models we found best to differentiate invasive OSCC and dysplasia from controls to determine if the expressions of these genes differ according to disease severity. The results will be used to formulate a subsequent proposal to investigate the association of these genes' expressions with local recurrence and/or second primary oral cancer. 3: Assess genome-wide loss of heterozygosity (LOH) and copy number aberration (CNA) in relation to clinical characteristics of OSCC, such as tumor size, stage and nodal metastasis in 100 0SCC patients. We will interrogate paired DNA from tumor cells isolated by laser capture microdissection and DNA from peripheral blood using Affymetrix 6.0 SNP arrays. We hypothesize that some LOH and/or CNA may be associated with stage and metastasis, and ultimately hope to determine the degree to which a combination of stage, LOH/CNA and gene expressions can better predict metastasis and/or survival than stage or gene expression, alone or in combination. HPV plays an important etiologic role in some oropharyngeal cancers, we plan to determine and incorporate HPV genotypes in the analyses and interpretation of the study findings.