Epidemiology Program

Oral Cancer: Molecular Profiles & Clinical Outcomes

PI: Chu Chen PhD

Oral and oropharyngeal squamous cell carcinoma (OSCC) is the sixth most frequently diagnosed cancer worldwide. In the U.S., approximately 30,300 new cases were diagnosed in 1999. The 5-year survival of patients with OSCC -about 50%- has not improved in the past two decades. Survivors often suffer orofacial dysfunction, disfigurement, and psychological stress. A major impediment to the effective management of OSCC patients is our limited ability to predict the natural history of individual lesions. While prognosis is to some extent correlated with anatomic site, tumor thickness, tumor grade, and lymph node involvement, there is much unexplained variability in the clinical course of OSCC patients. Physicians cannot reliably identify those patients (within a given tumor site, stage, grade, etc) who might benefit most from specific diagnostic and/or therapeutic interventions, such as screening for occult metastasis, or offering particularly aggressive surgical or radiation therapy. Studying tumor characteristics via examination of genomic scale gene expression may improve our understanding of tumor behavior. Our goals in this study are to determine whether: 1) The prognosis of OSCC patients is associated with tumor gene expression profiles; 2) OSCC tumor characteristics known to be related to patient outcome - such as stage and nodal status - are associated with tumor gene expression profiles; and 3) Tumor gene expression differs among invasive OSCC, pre-neoplastic oral squamous cell lesions, and normal oral tissue. We will enroll patients who are newly diagnosed with invasive OSCC or oral dysplasia, and patients with no current or past history of these conditions, at the University of Washington Medical Center and its two other affiliated medical centers. At the time of diagnosis, we will obtain from each patient fresh tissue from biopsies and resections and baseline demographic and baseline lifestyle information through an in-person interview. Patients will be followed prospectively to identify clinical outcomes through review of medical records, patient interviews, and linkage to the National Death Index and Social Security Death Index. We will identify genes whose expression profiles are associated with clinical and pathological endpoints of interest by 1) interrogating pathologically verified OSCC tumor tissues, dysplastic lesions, and normal tissues using Affymetrix U133A oligonucleotide arrays for 12,000 genes; 2) analyzing the array data with a modified linear regression model and Bonferroni correction while adjusting for demographic, lifestyle, and comorbidity variables; and 3) confirming gene expression by SYBR (r) Green I quantitative RT-PCR/melting point dissociation curve analyses. The results of this study have the potential to influence clinical care of oral cancer and oral dysplasia in that the molecular signatures we identify may permit physicians to 1) screen oral lesions and surgical margins of tumors for malignancy, 2) screen OSCC patients with negative cervical nodes for occult metastasis, and 3) identify patients with potentially poorer outcome who might benefit from more aggressive treatments. This study may also uncover genes and pathways that are important for oral cancer development, and potential molecular targets for chemoprevention and therapeutics.