Risk Prediction for Common Diseases

• Author: Paula W. Yoon
• Position: ScD, MPH, Epidemiologist, Office of Genomics and Disease Prevention
• Pages: 33-42

Page33

ScD, MPH, Epidemiologist, Office of Genomics and Disease Prevention, Centers for Disease Control and Prevention (CDC). This work is based upon a live presentation made on February 5, 2004 and does not eces rily reflect events and changes thereafter.

My topic for discussion today is whether the promising field of genomics is going to allow us to more accurately predict who in the population is at future risk for common chronic diseases. Before attempting to answer this question, I will provide some general background and will then give some specific examples in the area of heart disease, diabetes, and obesity.

I work in the Office of Genomics and Disease Prevention at the Centers for Disease Control ("CDC") and, like most of CDC, our focus is on disease prevention. Our office has a crosscutting role, working with all the programs at CDC to help them integrate genomics into what they do, whether it is research, policy, or practice.

The year 2003 was certainly the year of the human genome. We celebrated the completion of the human genome project and the fiftieth anniversary of Watson and Crick's discovery of the double helix structure of DNA.¹ One headline claimed, "DNA has changed the world: But now what?" That is the big question.

As scientists, we are pretty good at collecting DNA. We can get DNA from all kinds of body fluids and parts; it is amazing where we can find DNA.² Our technology for analyzing DNA, which now includes whole genome scans and chips, is also improving immensely, rapidly becoming faster and cheaper.³ But what does this mean for the health of individuals, families, and communities?

The work that the scientific community is doing can be thought of on a continuum, a translation continuum from gene discovery to tools and processes that can be used to prevent and treat disease. Along that continuum, there are various research methods that are used to study genomics. First, there are family studies or linkage

This article is not copyright protected under 17 U.S.C. 105 (2005). Page34 studies, which look for genetic differences within small groups. Next, there are population studies of different kinds--case-control studies, cohorts, and so forth. When established associations lead to new genetic tests or tools, there are clinical trials. Finally, after studying a test or tool under controlled circumstances, we need to evaluate its validity and utility in real circumstances. This last step--which should occur before it is introduced for widespread clinical or public health application--is often skipped or glossed over. I am going to come back to this validation gap at the end of my talk. Certainly, the use of emerging technology like whole genome scans and haplotype mapping ⁴ make the field of genomic research very exciting. The promise is there but we have a long way to go from gene discovery to treatment and prevention of disease.

When genomic applications, like genetic tests, are carried successfully through the research continuum what are their potential uses for treating and preventing disease? There are basically four areas of potential application. First is the possibility of guiding drug therapy. Pharmacogenomics, or individualized medication based on genetically determined variation in effects, is probably one of the most promising areas in the field of genomics today. There is individual variation in the enzymes that metabolize, absorb, and transport pharmaceuticals. There has also been some success in identifying genes that cause some individuals to have extremely positive or extremely negative effects from drugs. Another related example is the use of DNA probes to identify pathogens. For example, in the diagnosis of meningitis,⁵ DNA probes can be used to determine fairly rapidly whether the illness is due to bacteria or a virus and, specifically, which pathogen. This allows for quicker application of the most effective therapy.

A second way in which genomics can be used to prevent disease is by modifying the environment. If we can identify individuals who are more susceptible to disease because of their genetic make-up, it may be possible to reduce their risk through individual behavior changes, such as diet, exercise, or smoking cessation. Environmental modification can also occur through community interventions, such as spraying for mosquitoes or Page35 having more sidewalks in communities so people can walk to school and work. Workplace interventions are another possibility. For example, if we know that within a particular workplace certain people are more susceptible to substances like chemicals and pesticides, exposures levels can be lowered for the safety of everyone.

Gene therapy is another potential option for treating disease, but this is still experimental. There has been some success with using gene therapy to treat single gene disorders, such as primary immune deficiency, but results are mixed.⁶ The National Institutes of Health ("NIH") is funding a number of studies in gene therapy research to develop treatments for cancer, heart disease, and AIDS. This is an area of research that looks promising, but we have a long way to go.

A fourth avenue for disease prevention is offering targeted screening and interventions based on increased susceptibility.⁷ For example, a person at risk for hereditary colorectal cancer may be encouraged to be screened earlier or more frequently than is normally recommended and may benefit from more intensive screening methods, such as colonoscopy. Presymptomatic medical therapies may also be helpful, whether they are as simple as taking an aspirin a day or a prescribed medication.

These four main areas--drug therapy, environmental modification, gene therapy, and targeted interventions--are the key strategies we think of when we talk about using genetics and genomics to help us treat and prevent disease.

Where are we today on the research continuum for preventing common chronic diseases like heart disease, diabetes, and obesity? The last estimate from the Human Genome Project ("HGP") was that humans have between twenty and twenty-five thousand genes,⁸ as well as millions of variants of these genes. Common chronic diseases result from the interactions of multiple genes with multiple environmental and behavioral factors...