Long-Anticipated Gene Search Technique Is Now Powerfully Real

In the wake of national-headline-making reports about how inheriting particular bits of DNA can increase one's risk for diabetes and breast cancer, UCSF's John Witte, PhD, stands ready to try his hand at using a similar scientific strategy to track down genes that affect prostate cancer risk. Better tools to gauge individual cancer risks could lead to saved lives. Individuals at high risk could be screened for cancer more often, and could take actions to lower their risk. Cancer researchers have long probed human DNA to find variations in specific genes that contribute to cancer risk in individuals who inherit them. Now they are turning to a newer, comprehensive approach called a genome-wide association study - GWAS for short. Using GWASs, researchers are taking advantage of advances in technology and a growing knowledge of the human genetic code to search all human chromosomes for DNA variations that might affect disease risk. Leading Prostate Cancer Genetics Expert Witte has already led his share of more focused studies. He, like many other researchers, often studies "candidate" genes that encode proteins already known or suspected to play a role in biochemical processes that become altered in disease. In the last few years, he has published several papers which suggest that some of these DNA variations do indeed influence prostate cancer risk. But for his newly proposed study, Witte is not limiting his gaze to genes already identified as plausibly playing a role in prostate cancer. "We're no longer putting our bets anywhere," he says. "We're going to look everywhere." Witte also has prior experience searching across all chromosomes. He led a multi-institutional study in which research collaborators used earlier-generation probes to home in on a chromosomal hot spot where DNA variation helps predict the severity of already diagnosed prostate cancer. The proposed GWAS would be nested within a larger research endeavor, the Kaiser Permanente California Men's Health Study, led by Bette Caan, DrPH, and Stephen Van Den Eeden, PhD. The Kaiser Permanente project began five years ago, and aims to track health outcomes in more than 84,000 men. In collaboration with the Kaiser Permanente investigators, Witte plans to examine the DNA of 2,000 men who have been diagnosed with prostate cancer since the Kaiser Permanente study began, and to compare it with DNA from a similar group of healthy men from the same, larger study pool. Just as with candidate gene studies, follow-up of the GWAS with research on different populations of men would be required to confirm any new associations discovered between genetic variations and prostate cancer risk. Other groups have also begun using GWASs to study prostate cancer. In a GWAS on men from Iceland, and in a separate GWAS on US men by collaborating research groups from several universities, researchers tentatively identified scores of genetic variations among those men that were associated with the disease. Impact of DNA Variations Differs Among Ethnic Groups Follow-up studies published this May confirmed a role for several of these variations. A few occur commonly among African American men. New estimates indicate that this distinctive DNA may account for a significant amount of their 60 percent greater prostate cancer incidence, compared with white men.
John Witte

John Witte

The DNA variations identified or confirmed in those follow-up studies are not in genes, so they do not encode the blueprints for protein. They likely would have been missed by researchers interested only in focusing on particular genes. DNA outside of genes often functions to influence whether genes are turned on or off, but the role of the DNA where these variations appear is currently unknown. The study of Bay Area Kaiser Permanente patients will encompass an ethnically diverse population, and has the potential to identify many new genetic variations that affect prostate cancer risk, according to Witte. The Kaiser Permanente data also contain valuable information about medical histories and environmental exposures for comparison. "Both nature and nurture are important," Witte says. "People differ in their environments. If you ignore that - and assume it's the same for everyone - you might decrease your ability to detect genetic effects." Payoff Follows Human Genome Project Success Neil Risch, PhD, who now heads the Institute for Human Genetics at UCSF and is co-chair of the Department of Epidemiology and Biostatistics, clearly illustrated the potential value of the GWAS approach a decade ago in an oft-cited article in the journal Science. But GWASs only became a practical option after the completion of the Human Genome Project in 2003. That nationwide, $3 billion, 13-year collaborative effort involved hundreds of scientists and technicians working at many sites. Completed on schedule and within its budget, it provided the spelling for the entire genetic code - the genome - of a human. But even though we all have the same genes, ordered in the same way, our genomes - except for those of identical twins - are all unique in small ways. The Human Genome Project also sped the identification of millions of commonly inherited variations in DNA - as distinct from rare mutations. These variations are the foundation for human genetic diversity, including diversity in the genetic susceptibility to various diseases. The variations usually amount to single-letter differences in the genetic code. They are called single-nucleotide polymorphisms, or SNPs - pronounced "snips."

The starting sequences of the human genome. From the NHGRI Understanding the Human Genome Project CD-ROM. (See larger)

In the follow-up to the Human Genome Project, a global effort called the International HapMap Project, researchers collaborated to probe DNA in four diverse human populations. They identified most of the estimated 10 million SNPs that are present in roughly 5 percent or more of the individuals within a population. Coming to Grips with SNPs Biotech companies are providing ever-more informative generations of "SNP chips" to researchers. A SNP chip is really hundreds of thousands of lab tests arrayed in a grid on what is essentially a glass microscope slide. There's a reference SNP at every point within the array. If one of the reference SNPs matches a SNP from a research subject's DNA, a signal is generated, and is read and recorded using now-automated machinery. SNP chips give researchers unprecedented power to track human genetic variations. Certain DNA sequences, typically containing several SNPs, are likely to be inherited together without being broken up as they are passed down through generations. These blocks of DNA are commonly referred to as haplotypes. Because the SNPs on chips may be chosen so that they are optimally spaced across the entire genome, like mile markers on a highway, they can do a good job of tracking naturally occurring haplotypes in human populations. Researchers can determine which haplotypes are identical or different among individuals in any population, and how inheriting particular bits of DNA might be associated with a higher or lower disease risk in those populations. The latest chips boast upward of half a million to almost a million SNPs, covering the genome in unprecedented detail. "The field has changed," Witte says. "Now you can simultaneously measure enormous numbers of SNPs, which you couldn't do just a few years ago."
 a DNA sequence

From a DNA sequence that contains a gene of interest, the DNA strand that codes for the protein is called the sense strand because its sequence reads the same as that of the messenger RNA. The other strand is called the antisense strand and serves as the template for RNA polymerase during transcription. From the NHGRI Understanding the Human Genome Project CD-ROM. (See larger)

Witte will be collaborating with the Kaiser Permanente researchers, Risch and his UCSF neighbor, Pui-Yan Kwok, MD, PhD, who participated in the International HapMap Project. Kwok's lab group can now analyze nearly 100 samples per week, meaning that data for the new GWAS could be collected in less than a year. Benefits of Knowing Risk Will Grow By learning how cellular and physiological processes are affected by DNA discovered to be associated with disease, researchers hope to gain new insight into disease - and how to prevent and fight it. The ability to better identify cancer risks in healthy individuals also can offer more immediate health benefits when screening and prevention options are affordable and available. Cancer screening often saves lives because tumors are detected at an earlier, more treatable stage. But screening tests can also have a downside. The results are not always an accurate indicator. Results from mammograms or prostate-specific antigen (PSA) blood tests that ultimately turn out to be "false positives" trigger anxiety and require invasive workups before breast or prostate cancer can be ruled out. False positives diminish and overall cancer detection rates increase when the population being screened for disease is at higher risk. That's one reason researchers want to identify new risk factors for cancers and gauge their importance. Use of these genetic risk measures in medicine and public health programs would require an accurate, cost-effective test, as well as public trust in legal safeguards to prevent health insurance discrimination based on test results. The benefits of being able to better identify individuals at high risk will grow as additional preventive measures become available. Many women at elevated risk for breast cancer, for instance, already take drugs that lower risk by modifying estrogen's hormonal effects. GWAS Versus Linkage Prostate cancer differs from breast cancer in that until the most recent GWASs, no single DNA variant in prostate cancer had been identified that exerts a profound influence. In contrast, more than a decade ago researchers determined that some breast cancers are due to rare, inherited mutations in genes called BRCA1 and BRCA2. While the average lifetime risk for breast cancer among US women is about one in eight, many of the BRCA mutations that have been identified cause cancer in a majority of women who inherit them. Rare, potent, disease-causing genes like the BRCA mutations usually are tracked down within affected families within and across generations, using a technique called linkage analysis. The technique takes advantage of - and its power is limited by - the natural exchange and splicing of DNA that take place between the two members of each chromosome pair during the cell divisions that give rise to sperm or egg cells. A GWAS is not considered a cost-effective way to search for rare mutations. But when looking for common DNA variations that have less profound effects - a doubling of cancer risk, for example - a GWAS often makes more sense than linkage analysis, Witte says. GWASs also have recently proved fruitful in studies focused on families with a high incidence of cancer. BRCA mutations account for perhaps one-quarter of breast cancers that run in families. In three more studies published this May, independent teams of researchers reported the use of GWASs to identify six new sites of DNA variation along the genome that may account for much of the previously unexplained familial breast cancer.
News and Views: Multiple Prostate Cancer Risk Variants on 8q24 John S. Witte Nature Genetics 39:579-580 (2007) Full Text | Full Text (PDF)
Editorial: Genomics of Common Diseases Nature Genetics 39:569 (2007) Full Text | Full Text (PDF)
Related Links: SNP Chip Makers: Illumina, Affymetrix Cancer Genetic Markers of Susceptibility (CGEMS) Project National Cancer Institute Database of Genotype and Phenotype (dbGaP) National Center for Biotechnology Information Genetic Information Nondiscrimination Act of 2007 International HapMap Project The Human Genome: Genetics of Cancer
Osher Lifelong Learning Institute, UCTV, July 10, 2006