To find common genes that confer modest amounts of risk, researchers have begun to survey the human genome for DNA variants such as single nucleotide polymorphisms (SNPs) that are associated with some common cancers. Genome-wide association studies are underway for common diseases, including diabetes. Last year, NIH launched two initiatives to support such efforts.
The NCI-sponsored Cancer Genetic Markers of Susceptibility (CGEMS) project is using high-throughput technology to identify susceptibility genes for breast and prostate cancers. The genotyping data are being made available online so that other investigators can have access to the information.
"The hope is that genome-wide association studies will be a discovery tool for identifying genes that no one knew were involved in cancer," said Dr. Margaret Tucker, director of the Human Genetics Program in the Division of Cancer Epidemiology and Genetics (DCEG). "These genes could point to novel environmental risk factors and biological mechanisms in the disease."
Some research groups have been establishing consortia and pooling resources to assemble the large number of individuals required to detect modest effects of genes on cancer risk and to replicate preliminary findings.
The NCI-sponsored Cohort Consortium is an international collaboration of researchers responsible for about 25 population cohorts involving 2.6 million individuals. The Cohort Consortium provides an integrative framework for studies of specific cancers to systematically evaluate biomarkers of susceptibility and disease.
A recent study by another group, the Breast Cancer Association Consortium, suggests that researchers can confirm or refute associations reported previously in the scientific literature by pooling data. The consortium used data on 33,000 women in 13 countries to confirm that a SNP in the gene CASP8 may offer modest protection against the disease.
The finding does not have immediate implications for prevention. But it suggests that researchers could use the same approach to identify panels of genetic variants that collectively influence a woman's risk, said Dr. Montserrat Garcia-Closas of DCEG.
Identifying SNPs linked to disease in genome-wide scans is an important first step, but additional work is usually necessary to identify the genes involved. "The data from these scans should be a rich discovery resource for the whole community and should stimulate new areas of research," said Dr. Tucker.