 |
 |
 |
FDA Home Page | Search FDA Site | FDA A-Z Index | Contact FDA
| Email this Page To a Friend |
 |
By Michelle Meadows
Genes determine the color of our eyes and shape of our bodies. Genes also determine our susceptibility to disease and how we respond to medicine.
Researchers believe that each person has about 35,000 genes. The complete set of genes together is known as the human genome, commonly referred to as "the instruction manual" for how the body works. Each gene carries instructions for making proteins, which direct the body's cells and functions.
Most cells have 46 chromosomes--23 from each parent. Chromosomes contain thousands of genes, which are made up of deoxyribonucleic acid (DNA), the chemical material that is inherited.
Genomics is the study of an individual's gene structure, including how the genes interact with each other and with the environment. Experts say genomics has the potential to revolutionize the practice of medicine. That revolution, called "personalized medicine," includes the use of genomic information to improve the diagnosis of disease, as well as the prevention and treatment of disease.
An example of a preventive approach is when a genetic test predicts which diseases an individual is likely to develop. For instance, people who have certain mutations in the BRCA1 gene have a high risk of developing breast, ovarian, and possibly prostate, and colon cancers, according to the National Cancer Institute (NCI). Alterations in the BRCA2 gene have been associated with breast, pancreatic, gallbladder, and stomach cancers. An example of a treatment approach is when a genetic test determines whether a person is among the 10 percent of those for whom a particular drug is likely to work.
Felix Frueh, Ph.D., associate director for genomics in the Food and Drug Administration's Office of Clinical Pharmacology and Biopharmaceutics, says, "Personalized medicine tries to answer questions like: Why do some people get cancer and others don't? Why is cancer more aggressive in this person compared to that one? Why does this drug work for you and not me? Why does someone need twice the standard dose to be effective? And why do others need only half of the standard dose?"
"The goal of personalized medicine is to get the best medical outcomes by choosing treatments that work well with a person's genomic profile, or with certain characteristics in the person's blood proteins or cell surface proteins," Frueh says. Genetic information isn't usually meant to be used alone to make treatment decisions, but rather is used with other factors such as the patient's family history, medical history, clinical exam, and other non-genomic diagnostic tests.
The combination of drugs (pharmacology) with genomics is known as pharmacogenomics, the science that allows researchers to predict the probability of a drug response based on a person's genetic makeup. "It's about getting the right dose of the right drug to the right patient at the right time," Frueh says.
The science of pharmacogenomics has advanced significantly in the last five years, but it's still in its infancy and is mostly used on a research basis, says Larry Lesko, Ph.D., director of the FDA's Office of Clinical Pharmacology and Biopharmaceutics. "There are three main ways that pharmacogenomics is applied," Lesko says. "The first is to help predict the appropriate dose of a drug. The second is to target therapy to a subset of a disease. This means picking the most effective drug for the disease subset. And the third is to test viral genomics, such as in selecting treatment for HIV based on resistance."
The usual doses of drugs work well for most people. They are sometimes based on weight, age, and kidney function. But for someone who metabolizes a drug quickly, the typical dose may be ineffective and a higher dose may be needed. By contrast, someone who is a slow metabolizer may need a lower dose; the typical dose could cause toxic levels of the drug to build up in the blood.
When we take medicine, it moves through our body, gets broken down by drug-metabolizing enzymes, and interacts with countless proteins. "Genes regulate drug metabolism," Frueh says. "Differences in the sequence of a gene can cause differences in enzyme activity, which is a result of enzymes appearing in various forms in individuals. This is why different people process the same drug differently."
Mary Relling, Pharm.D., chairwoman of pharmaceutical sciences at St. Jude Children's Research Hospital in Memphis, Tenn., says that for children with leukemia, getting the dose wrong can mean the difference between life and death. "We want to keep harsher treatments away from children whose bodies can't tolerate them or don't need them," Relling says.
For example, St. Jude's routinely conducts a genetic test for defects in the enzyme thiopurine methyltransferase (TMPT). The defect prevents patients from metabolizing the anti-cancer drug 6-mercaptopurine (6MP). One patient may need a full dose; another, who has a mutation in the gene, may need less than 10 percent of that dose.
"Before this test, some patients needed blood transfusions or had to be hospitalized for infections, and we didn't know why," Relling says. "Now we can avoid some of the toxicity of the drug by testing all the children who come in and making dose adjustments based on their genetic test and response to therapy."
Targeted therapy, the second major aspect of pharmacogenomics, is also referred to as "tumor genomics," Lesko says. Tumors have different genomic variations, and genomic tests are helping doctors to identify cancers that are likely to respond to a particular treatment. Lesko cites the drugs Gleevec (Imatinib) for chronic myeloid leukemia, Tarceva (erlotinib) for lung cancer, and Herceptin (trastuzumab) for breast cancer as examples of targeted therapy.
Both Gleevec and Tarceva interact with enzymes called tyrosine kinase inhibitors. Turning off these enzymes prevents the growth of cancer cells. Herceptin targets tumors that produce excess amounts of the HER2 protein, which is produced by the HER2 gene. Overexpression of the HER2 protein causes a higher rate of cell growth. Before Herceptin is used, tumors must be tested to evaluate the amount of HER2 protein.
The third aspect of pharmacogenomics includes testing for drug resistance, Lesko says. For example, the HIV virus genome is always changing, and resistance testing can help doctors choose the drug that will best match the virus and suppress it. The TRUGENE HIV-1 Genotyping Kit is cleared by the FDA to detect genetic variations that make the HIV virus resistant to some anti-retroviral drugs. If drug resistance is discovered, a doctor can decide to try another treatment option.
The main benefit of pharmacogenomics for consumers is the availability of drugs that have a greater chance of benefit in terms of treating illness, says Janet Woodcock, M.D., the FDA's deputy commissioner for operations.
"Consumers want effective treatment and minimal side effects," Woodcock says. "They want to know that they're getting the right drug and the right dose. There are some people who feel like they are always experiencing bad side effects from drugs, and in some cases, genetics plays a role. If we could find out who the susceptible people are so they can avoid the risk, we could target a drug more appropriately instead of removing it from the market."
The hope for the future is that through personalized medicine, doctors and patients will be able to make better-informed choices about treatment. Genetic information could lead them to decide which drug to use, whether to lower a dose, or whether closer monitoring of the patient for side effects is needed.
The most common approach to drug treatment now is that doctors give all patients with a given disease the same drug and an average dose, evaluate how it works, and then make adjustments as needed. "But with life-threatening illnesses such as cancer and heart disease and with drugs that can have serious side effects, getting it right the first time is critical," Relling says. "Rather than using a trial-and-error approach, we want to be able to analyze a patient's genetic profile and prescribe the best therapy and dose from the start."
With technology called gene expression profiling, researchers can determine which of the genes that activate or break down drugs are active in a particular patient. "This could show who is most likely to benefit from a drug and who may have a toxic reaction," says Relling, who, with colleagues at St. Jude's, has led some of the world's first pharmacogenomic studies in children with acute lymphoblastic leukemia (ALL). This type of cancer starts in the bone marrow, the soft, spongy inner part of the bones. The disease can move quickly into the blood or other parts of the body such as the liver, the spleen, and lymph nodes.
"Chemotherapy cures about 80 percent of ALL cases," Relling says. "But treatment fails in the remaining cases because the patients' cancer cells are resistant to chemotherapy drugs or because of fatal toxicities of chemotherapy. The hope is that cure rates can increase if doses are individualized to maximize cure or minimize adverse effects."
Relling and her colleagues at St. Jude's routinely use genomic tests to differentiate between various subtypes of ALL, which respond to chemotherapy in different ways.
"We have genetic studies built into the treatment protocols as part of our research program," Relling says. "We measure how much of the drugs are in the children's leukemic cells and their blood. We perform MRI imaging of their hips to see if there is any bone toxicity, and we follow side effects and leukemia status for two and a half years while children are on chemotherapy. We put together all of this detailed information with genetic testing and identify which genetic profiles predict side effects or relapse. We not only look forward, but we also look back at data we've collected from years ago. We started collecting DNA from patients in 1986 because we anticipated that this genetic revolution was coming."
This type of research could make drugs for many diseases more effective, including heart disease, diabetes, depression, and asthma, says Michael Caldwell, M.D., head of the Personalized Medicine Research Center at Marshfield Clinic in Marshfield, Wis. Caldwell and his colleagues have investigated the role of genetics in a number of drugs, including the anti-clotting drug Coumadin (warfarin).
"You can give the standard dose of warfarin and some people are just fine, while others will have significant complications such as bleeding into the brain," Caldwell says. "We confirmed work by others that if you identify a patient's genetic make-up, we can get a better handle on the appropriate dose much earlier."
Caldwell says he believes that the field of personalized medicine will continue to grow in the interest of patient safety. "I think we will gain a better ability to predict adverse reactions earlier in drug development." And decreasing adverse events and increasing successful therapy could lower the cost of health care, says Allen Roses, M.D., senior vice president for genetics research at GlaxoSmithKline (GSK).
"Medicines are often marketed broadly because the patients likely to benefit from a particular drug could not be separated from the patients who would not respond well," Roses says. "But it could be more economical to make informed decisions based on inherited characteristics and the variability that exists in the population in order to produce safer and more effective drugs."
Roses says the potential of pharmacogenomics for drug companies lies in the discovery stage--being able to create new drugs based on genetic information. A better understanding of genetics and disease helps identify new targets for drugs. This could bring drugs to the market sooner.
"If we find in clinical trials that 90 percent of people don't respond to a drug, we can look at groups of nonresponders that have the same genetic characteristics and conduct further research," Roses says.
GSK routinely collects DNA from patients who provided informed consent for pharmacogenomic studies in its U.S. clinical trials. "This will help identify factors that may predict drug response," he says. "If only some people are experiencing adverse events and DNA is available for those patients that will help tell us something about how to lower the risk of side effects."
"In clinical trials, drugs that may be harmful to only a small number of patients might be able to be used if we find out which genes are associated with the adverse responses," Roses says. "The drug could be approved for people with a low risk of adverse events."
In 2002, the FDA held a workshop for industry on pharmacogenomics in drug development and regulatory decision making. "The agency has taken a leadership role by initiating discussions about pharmacogenomics and gathering input on how the FDA can best promote the science," Woodcock says.
In March 2005, the FDA released a guidance on the agency's current thinking about pharmacogenomics and on data submission. The FDA considers pharmacogenomics to be a major opportunity on the critical path to new medical products. For now, most pharmacogenomic data are of an exploratory or research nature and are not required under existing regulations; however, voluntary submissions are encouraged.
"This way, industry can meet informally with the FDA and talk freely without concern that the data will be used inappropriately for regulatory decision making and at the same time drug developers will learn more about the agency's expectations," says Frueh. The FDA is working on other guidances on how to develop genomic tests and drugs together and on the quality of DNA analysis.
"From FDA's perspective, genomic data has the potential to help us understand how a drug works and for whom, which helps in assessing the risk-benefit ratio," says Yvonne Dragan, Ph.D., director of the division of systems toxicology at the FDA's National Center for Toxicology Research (NCTR) in Jefferson, Ark. Scientists there are active in pharmacogenomic research. For example, FDA experts found that women with breast cancer who have a slow version of the gene SULT1A1 have lower survival rates on a typical post-surgery regimen of tamoxifen and may need different doses of the medicine.
"The FDA also plays a critical role in making the tools available to put the science into practice," Dragan says. In December 2004, the FDA cleared the AmpliChip Cytochrome P450 Genotyping Test. Doctors can order this test to gain information on whether the patient has mutations in a gene that's active in metabolizing many types of drugs, including antidepressants, antipsychotics, beta-blockers, and some chemotherapy drugs.
In August 2005, the FDA cleared the Invader UGT1A1 Molecular Assay, which also detects variations in a gene that affects how certain drugs and drug metabolites are broken down and cleared by the body. Variations in the gene can influence the patient's ability to break down the major active metabolite of Campostar (irinotecan), a drug for colorectal cancer. The inability to break the metabolite down can lead to increased levels of it in the blood and a higher risk of side effects.
"The FDA's role in personalized medicine is one of bringing balance to an evolving science in a way that reaps the benefits, but also doesn't inhibit its growth," Lesko says. "It's a flexible approach, and we have been engaging industry in continuous dialogue, issuing guidances as the science matures, and continuing to focus on the impact pharmacogenomics can have on public health. We want to get the tools out there so that doctors can make informed decisions and so they will be able to interpret what value and meaning genomics brings for their patients' medical care."
The FDA will be working to advance its capabilities to analyze and interpret genomic data that come into the agency, and to communicate the relevant information in an understandable way in drug labels, Lesko says. "We will also be working to coordinate the efficient and timely review between centers in the agency as drugs and tests are co-developed. More and more, genetic tests will be used along with drugs."
A major challenge, Lesko says, is that there is a general lack of familiarity with pharmacogenomic data because the science is so new and is constantly evolving in new ways. "I think it will require a new way of thinking about individualization, and some people will be resistant to change given the challenges of understanding the information," he says.
Relling agrees. "Most clinicians will be resistant to the routine use of pharmacogenomics in medical care," she says. "Incorporating pharmacogenomics into prescribing decisions will represent a major change for the health care community."
"The hope for the future is that every patient will have a blood sample and we can extract DNA and do a test to get a genetic profile. Then we could use the profile along with clinical information to make care better," she says. "But it's not going to be that simple. It's going to make practicing medicine more complicated. Most of the world uses a trial-and-error approach. So changing will come at some cost to society."
Experts say that incorporating pharmacogenomics into everyday medicine will also take time. It is happening in mostly academic medical institutions like St. Jude's. Even though the Human Genome Project has sequenced a map of the body's genes, that doesn't identify all the targets in drug discovery, Relling says. "Many more studies will be needed to identify which gene variations are related to drug response and to prioritize them." The Pharmacogenetics Research Network of the National Institutes of Health is a nationwide collaboration of scientists committed to studying the effects of genes in response to a variety of medicines, including antidepressants, chemotherapy, and drugs for asthma and heart disease.
"In cancer, we're also dealing with diseases that require at least 10 years of follow-up so we can see the long-term outcome for patients," she says. "The public needs to understand that it may take years before we know whether individualizing doses based on genetics improves the long-term outcome of cancer patients because of the nature of the disease. Careful collection of clinical information and studies with large numbers of patients will be required," Relling says.
"In the future, legal assurances will be needed so that insurance companies can't use genetic information to discriminate against people who have a genetic risk for health problems," she says.
Caldwell says there also may be reluctance on the part of doctors and researchers to pay for and conduct genetic tests until someone shows that these tests make a difference. "Multiple genes can influence patients' response to the drug, as well as other drugs given at the same time, and various environmental factors," Caldwell says.
Experts say that another challenge will be whether drug companies decide that there is enough financial benefit from therapies that help only small segments of people.
"We're really at the dawn of this," says Caldwell. "We will make new discoveries about genes that are important in drug treatment and we will see more diagnostic tests. The next step now is proof of concept--proving the difference that personalized medicine makes."
Genomics at the FDA
www.fda.gov/cder/genomics/
Personalized Medicine Coalition
www.personalizedmedicinecoalition.org
The FDA's National Center for Toxicological Research
www.fda.gov/nctr/
The CDC's Office of Genomics and Disease Prevention
www.cdc.gov/genomics/
NIH Pharmacogenetics Research Network
www.nigms.nih.gov/Initiatives/PGRN/