| BACKGROUND: Colorectal cancer is the second leading cause of cancer death in the western world with over 4200 deaths per year in The Netherlands. Today, secondary prevention is the most realistic approach for reducing this high number of colorectal cancer deaths. At present, two options for secondary prevention exist: One is fecal occult blood testing which is easy and well tolerated, but has a low sensitivity and low specificity. The other is colonoscopic screening. This has a ± 90% sensitivity for detecting the precursors of colorectal cancer, i.e. adenomas, but since only ± 5% of all adenomas ever progress to cancer, it has a low specificity in terms of cancer prevention. Therefore, a clear need exists for more sensitive and specific tests for colorectal cancer, which preferably are easy to apply and well tolerated. First attempts for genetic testing on tumor cells shedded in the feces are promising, but still have limited sensitivity and specificity. Including additional genes associated with colorectal adenoma to carcinoma progression in these tests is likely to improve sensitivity and specificity of fecal gene tests. We recently have shown that areas on chromosome 8q22-23, 13q21-31 and 20q13 probably harbor such (onco)genes. Identification of these genes would allow to use them as targets in fecal gene testing. An improved genetic fecal test will result in a better selection of candidates for colonoscopic intervention, which will make the use of available resources and patient care more efficient. In addition, knowledge of these oncogenes will open up possibilities for intervention with tumor progression by chemoprevention. The regions at 8q22-23, 13q21-31 and 20q13 still contain hundreds of genes, reason why we first need to zoom in. Today, micro-array comparative genomic hybridization of tumor tissue is the single most efficient tool to narrow down areas of chromosomal gain of several megabases to regions as small as e.g. 100 kB and is successful in our hands. In combination with surveys of human genomic databases and expression array analysis, we reckon this scenario the most promising for identifying oncogenes important for colorectal adenoma to carcinoma progression. PURPOSE: To further dissect the amplified chromosomal loci 8q22-23, 13q21-31, and 20q13 to identify novel putative oncogenes important in tumor progression of colorectal adenomas, using sporadic malignant polyps as a model. PLAN OF INVESTIGATION: Phase 1: Determine amplicon boundaries for the amplifications at 8q22-23, 13q21-31 and 20q13. First, we will dissect these loci by micro-array CGH with targets spread at medium (500kb) resolution. This will narrow down the common region of overlap to 1-3 Mb. Using an array with a contig of overlapping target clones, exact breakpoints can be identified yielding regions of approximately 100 kb harboring the putative oncogenes. One hundred DNA samples obtained from micro-dissected pre-invasive parts and invasive parts of colorectal tumors (already analyzed by classic CGH) are available for this purpose. This strategy is feasible and has already been proven to be successful in breast and gastric cancer. Phase 2: Identification of the most likely candidate oncogenes. By extensive surveys of human genomic databases, cDNAs of known genes and ESTs located within the amplicon will be selected as targets to be used in expression micro-arrays. Overexpression at the mRNA levels of genes within the ampicon will point at putative ongogenes. In addition, expression patterns of a number of invasion-related factors known from literature, such as matrix degradation enzymes, cell adhesion proteins, and angiogenesis and stroma-inducing factors will be analyzed using the Dutch Cancer Society (DCS) 2.5k expression micro-array which includes approximately 150-200 of these genes. For these expression experiments, snap frozen tissue samples from 50 adenomas and 50 carcinomas from colectomy specimens will be used. With this approach we limit the size of the arrays to save time and produce manageable amounts of data. POSSIBLE RESULTS / RELEVANCE FOR CANCER RESEARCH: Given the magnitude of the problem and the present limitations in our possibilities to reduce colorectal cancer death, any progress in identifying genes clearly associated with progression of adenomas to colorectal cancer would mean a success. The proposed project will identify putative oncogenes in chromosome regions 8q22-23, 13q21-31, and 20q13. These oncogenes, together with APC and p53, then could be used in fecal tests for more specifically detecting patients with high-risk colorectal adenomas as candidates for colonoscopy. In addition, these new oncogenes may serve as targets for early intervention of tumor progression by chemoprevention. |
