Progression of cancer is a multi-step process. Mutations occur in the genome of the precancerous cells and accumulate until the growth of these cells goes unchecked. Identification of genomic regions that undergo mutation will lead to the discovery of genes, oncogenes and tumor suppressors responsible for tumor progression. We have developed a genomic microarray-based method, representational oligonucleotide microarray analysis (ROMA), to identify regions in the cancer genome that have undergone copy number fluctuation. This method is highly sensitive and quantitative and can detect loss of a single allele.

We are currently studying two tumor types, ovarian and pancreatic cancer, with two goals in mind. First we will be analyzing a large set of tumors to identify the genomic regions that are mutated frequently. These regions will be searched for gene candidates and these candidates will be functionally characterized. Second, changes that occur in a tumor may constitute a genetic fingerprint of that tumor. We plan to identify similarities in genetic fingerprints using informatics methods. This data will be compared to information obtained from clinical collaborators to correlate tumor genetic fingerprints and disease status, ultimately leading to improved diagnostic and treatment strategies.