Head and neck squamous cell carcinoma (HNSCC) is the fifth most common malignancy worldwide, representing a major international health problem.  These tumors constitute an anatomically heterogeneous group of neoplasms that share in common a causal association with tobacco and alcohol exposure.  The clinical course of these neoplasms is difficult to predict based on established prognostic clinicopathological criteria.  Given the genetic complexity of head and neck cancers, it is not surprising that correlations with individual genetic abnormalities have also been disappointing.  There is an increasing necessity for better classification of head and neck tumors in order to provide prognostic information to improve patient care.  Two projects ongoing in my lab are concerned with the elucidation of the molecular signatures of these tumor cells using high-throughput technologies:

A.  Head and Neck Tumor Classification Project
      Previous studies by our group have demonstrated that there is a link between the gene expression profiles of patient HNSCC tumor samples and the clinical outcome for that patient, suggesting that custom profiles can be developed to predict outcome (Belbin et al., 2002).  My lab is now conducting a prospective study utilizing a large patient population, along with microarrays containing 28,000 cDNA clones to identify new prognostic molecular markers for HNSCC.  Gene expression data from these experiments are being used as a “training set” to identify changes that correlate with histopathology, metastasis, clinical outcome and survival in these patients.  We are also testing whether or not our predictions based on multiple gene assessments are a more powerful prediction tool than single molecular marker immunohistochemical assessment.  While surgery can cure early stage disease, multimodality therapy is of limited success in later stage HNSCC. Thus, new diagnostics that can predict tumor behavior including response to therapy will have high clinical impact.  The printing of a custom microarray containing these prognostic molecular markers would represent an innovative tool to improve clinical decision-making in the care of head and neck cancer patients.  Furthermore, our findings will lead to more unique gene discoveries and, in the future, new targets for anti-tumor drugs.

B.  Deciphering the Mechanisms of Epigenetic Gene Silencing in Head and Neck Carcinogenesis
      It is now well-known that the hypermethylation of CpG island promoter elements and the transcriptional silencing of corresponding genes have emerged as one of the most frequent epigenetic events in human malignancies.  Our group has begun a project to study the role of CpG island promoter hypermethylation in tumor progression.  First, our group has extended the use of microarray technology to include the global analysis of DNA hypermethylation patterns in cancer cell genomes using a microarray of 12,288 CpG island clones.  Such DNA hypermethylation events represent a stable tumor-specific marker occurring early in tumor progression and easily detected by PCR-based methods in a manner that is minimally invasive to the patient.  It is therefore intriguing to consider DNA methylation signatures as a marker for early detection of carcinogenesis in HNSCC, as well as a classifier and prognostic parameter in patients with latter stage disease.  Utilizing our database of CpG island sequences, we can map diagnostic and prognostic CpG island clones onto the human genome, thereby linking CpG-rich promoters of interest to known genes. 
      My group is also interested in the role of the DNA methyltransferase (DNMT) family of enzymes in methylation events in HNSCC tumorigenesis.  We are assessing the relative expression levels of DNMT isotypes in HNSCC primary tumor tissue and normal adjacent mucosa, as well as HNSCC cell lines.  We are utilizing a combination of genetic (RNAi) and pharmacologic inhibitors to elucidate the functions of individual DNMT isotypes in HNSCC cell lines.  The joining of methylation microarray and gene expression microarray data will shed new light on the establishment and maintenance of CpG island hypermethylation in carcinogenesis.  Furthermore, the identification of genes specifically re-expressed in response to DNMT inhibitors represents a powerful approach for the comprehensive identification of new tumor suppressor genes in head and neck cancer.