Genetic Testing For Cancer Presents Complex Challenges

HARD NUMBERS: Penn's Fergus Couch evaluated the "actual prevelance of BRCA1 mutations" among 263 women with a family history of the disease. In the early days of the Human Genome Project, attention focused on rare disorders that strictly follow Mendel's laws of inheritance, recurring in predictable patterns within families. With many of these genes identified and mapped as the finish line nears, emphasis has shifted to genes that cause more common ills, particularly cancers, and the public has become more aware of genetic testing. These conditions are transmitted in more complex ways; the complications are creating problems for technology transfer from the research lab to the clinic. EDUCATION NEEDED: Johns Hopkins' Gloria Petersen found that many health care professionals misinterpret the results of genetic tests. The unexpected complexity of genetic testing technology is threatening its acceptance by consumers. "It is a very difficult situation. New discoveries attract publicity, and the public hears about them and demands information and even tests from their health-care professionals-who may not know how to order a genetic test or understand its results," says Gloria Petersen, an associate professor of epidemiology at Johns Hopkins University. Petersen was part of a research group that recently found 31.6 percent of 177 physicians queried misinterpreted the results of a genetic test for colon cancer (F.M. Giardiello et al., New England Journal of Medicine, 336:823-7, 1997). To address these issues, an education session at the annual meeting of the American Society of Human Genetics, to be held October 28 through November 1 in Baltimore, will discuss "counseling for a genetic susceptibility to cancer." Health Insurance in the Age of Genetics http://nhgri.nih.gov Task Force on Genetic Testing final report ww2.med.jhu.edu/tfgtelsi/fedregister/ Myriad Genetics Laboratories Inc. http://www.myriad.com Oncormed Inc. http://www.oncormed.com The problem of genetic testing today is twofold, recent reports reveal. Health-care professionals need to update their knowledge of genetics, and the mutation analysis that is the backbone of diagnostic and predictive genetic testing requires clarification from further research into gene function. Genetic researchers map and identify disease-causing genes by studying families with many affected members, homing in on a DNA sequence that ill relatives share. Once a gene is found, detecting it becomes a diagnostic or predictive test-but the risks may not be the same for the general population as they are for the families who helped identify the gene. Most people are less likely to inherit the disease-causing gene variants (alleles) and, if they do so, may be less likely to actually become ill. Two recent studies highlight these subtle distinctions. Fergus J. Couch, a senior research investigator in the department of medicine at the University of Pennsylvania, and colleagues evaluated the incidence of the breast cancer gene BRCA1 among 263 women reporting a family history at their clinic. Only 17 percent of this self-selected group had the mutation, compared with 45 percent of individuals in the families used to pinpoint the gene (F.J. Couch et al., N. Engl. J. Med., 336:1409-15, 1997). "The study was set up to define the actual prevalence of BRCA1 mutations in the populations that attend these clinics. There is no point in quoting frequencies to patients if they have no bearing on the patients themselves," says Couch. Jeffery Struewing and coworkers at the National Cancer Institute and the National Human Genome Research Institute tackled the question of risk if BRCA1 appears in a person whose family does not have many cancer cases. They examined questionnaires and blood from 5,318 Jewish people of Eastern European descent (Ashkenazim) who volunteered from the Washington, D.C., area, three-quarters of whom reported no personal or family history of breast or ovarian cancer. The researchers found that 120 of the participants had the three most common disease-causing alleles of BRCA1 or BRCA2. To evaluate cancer risk, the researchers considered how many relatives of the 120 individuals had breast, ovarian, or prostate cancer (the types associated with these genes). That risk is 85 percent for a person in the high-risk families used to find the genes. This approach produced a risk of developing cancer of only 56 percent. Struewing, the principal investigator from NCI's genetic epidemiology branch, concluded that the risk of breast cancer is not uniformly high for all women who carry a BRCA1 or BRCA2 mutation. He reported that other genes or environmental factors, still unidentified, may modify an individual's risk of developing cancer. LIMITATIONS: Determining risk depends "on patient recall of cancer in the family, and that is not always completely accurate," says Myriad Genetics' Susan Manley. The media interpreted the results of the study as casting doubt on the efficacy of BRCA1 and BRCA2 testing. But some people in the genetics community point out limitations of the study. "It is based on patient recall of cancer in the family, and that is not always completely accurate," says Susan Manley, a genetic counselor at Myriad Genetics Laboratories Inc. in Salt Lake City. Yet a risk of 56 percent is still high. "In the general population, the risk is 8 [percent] to 10 percent. So this value of 56 percent among Ashkenazim is a five-fold increase," says Leslie Alexandre, vice president of corporate affairs for Oncormed Inc. in Gaithersburg, Md. UNCERTAINTY: Utah's Jeffrey Botkin notes that there are too many mutations in BRCA1 to state the risks associated with each variation. Because a gene is an informational molecule, it can be altered in many ways. Geneticists can tell if a deviant sequence causes illness only if they know what the gene's product does-which isn't yet so for the breast cancer genes. "There are hundreds of mutations that have been identified in BRCA1. It is not possible to state the risks associated with many of these mutations, and it is unclear whether many of the gene sequence changes are normal variants [polymorphisms] or variants that increase cancer risk," explains Jeffrey R. Botkin, an associate professor of pediatrics and biomedical ethics at the University of Utah in Salt Lake City. But this uncertainty creates a nightmarish atmosphere for testing. While many of the noncommercial labs offering BRCA1 testing screen only for the most common alleles, making results clear-cut, commercial lab Myriad sequences each patient sample, which costs $2,400. (Oncormed does direct sequencing only if common alleles are not initially detected. Cost ranges from $300 to $1,000 for allele-specific tests.) Direct sequencing yields three types of results: a truncating mutation, which produces a shortened protein; no mutation detected; or a missense mutation (base substitution). These three types apply to colon cancer testing as well as breast cancer testing. It was this distinction that confused the physicians who interpreted absence of the truncating mutation as a negative colon cancer test result. "It was an inconclusive test result, not a true negative," explains Petersen. "The fact that a mutation wasn't found in the protein truncation test didn't mean that the patient did not have another mutation." In some cases, Myriad can tell whether a missense mutation is harmless or not. If the mutation is present in affected family members, or if the literature reports such an association, then it is linked to increased risk. Sometimes testing a woman's father can establish that a missense mutation is harmless. If the cancer is on her mother's side, and her father has the missense mutation she inherited, then a link to cancer is ruled out. But more often, the lab report for a missense mutation states that the risk is "uncertain." In 1995, the National Institutes of Health-Department of Energy Joint Working Group on the Ethical, Legal, and Social Implications of Human Genome Re-search, known informally as ELSI, established a Task Force on Genetic Testing to set public policy in this emerging medical arena. The final recommendations, released last May, include: requiring institutional review board approval of predictive genetic tests; requiring test developers to submit data on clinical utility and validation to outside review; strengthening of training in genetics for medical students, medical residents, and medical specialists whose patients have diseases with inherited components, and of training programs for nurses, social workers, and public health workers; developing criteria for stringent scrutiny of predictive genetic tests; requiring hospitals and managed care organizations to provide evidence of competency before their employees can order genetic tests or counsel about them; and improving transmission of information on rare genetic diseases to physicians. - R.L. Paradoxically, even as medical organizations are educating their members in the fine points of 1990s genetics, some people are looking far ahead and are wondering whether genetic testing will one day be done at home. "I'd say that by the year 2010, personal computers will be able to analyze DNA, and home genetic testing will be a big part of preventative health-care," predicts Glenn E. McGee, an assistant professor in the Center for Bioethics at the University of Pennsylvania. CD TESTING: Gamera Bioscience's Alec Mian believes his company's compact disc player that performs genetic tests will eventually "be at the consumer level." Signs of home genetic testing are on the horizon. Gamera Bioscience Corp. in Medford, Mass., is developing a compact disc player that performs genetic tests. The revamped CD unit includes input ports for body fluid samples, internal capillary tubes to carry out tests, and reagent reservoirs, according to CEO Alec Mian. The "LabCD System" can perform several different tests simultaneously on one sample, enabling physicians to tailor tests to specific patients. The company will initially target test manufacturers as clients but plans to broaden the market to physicians and patients. "The final product will be incredibly easy to use, even at home, although we are not aiming at that market now. But ultimately this will be at the consumer level," Mian adds. Meanwhile, some individuals are jumping the gun on genetic testing. For example, until midsummer, a person identifying himself only as "Larry" ran a Web site introducing a company called "Genetic Testing, LLC" that enticed readers to send in a health questionnaire to see which genetic tests they should have. Objecting to such phrases on the Web site as "Who should be genetically tested? Anyone concerned about a genetic disease in their family," the National Society for Genetic Counselors sent a letter of protest, and the site vanished. ("Larry" did not return repeated phone calls from The Scientist seeking comment.) And Manley reports that a physician advertised in a Jewish newspaper, offering genetic tests to anyone interested. The Task Force on Genetic Testing (see accompanying story) in its public policy recommendations released last May, comes down hard on direct marketing of genetic testing to the public. "Great care must be taken that information on genetic tests presented directly to the public is accurate and includes risks and limitations as well as benefits. Consumers should discuss testing options with a health-care provider competent in genetics prior to having specimens collected for analysis. The Task Force discourages direct advertising or marketing of predictive genetic tests to the public." FUTURE TESTS: By the year 2010, home genetic testing will be a big part of preventative health care," predicts Penn's Glenn McGee. In contrast are Myriad and Oncormed, the commercial laboratories offering breast cancer gene testing. Bioethicists have criticized them for offering their tests to anyone who asks, what McGee calls "drive-through genetic testing." But both companies carefully instruct physicians on how to screen prospective pa-tients; offer ex-cellent educational materials for medical professionals and consumers; and report that most of their samples come from comprehensive cancer centers, not general practitioners with limited genetic knowledge. "There are different types of patients, and quite specific criteria as to who should be tested and who should not be tested. We only test high-risk patients, and the patient must be competent to provide informed consent," explains Alexandre. Both companies also assist physicians in locating genetic counselors or providing counseling themselves. Although the uncertainties of genetic testing await resolution, fear of loss of insurance coverage is easing. "The risk of insurance discrimination has been overstated," says Botkin. Manley, reporting from the trenches, agrees. "Myriad has a reimbursement assistance program to help get insurance coverage. Insurance companies are ready to cover testing but want protective measures in place so not everyone is tested." Legislation is already heading off potential insurance problems when the Human Genome Project spawns many new genetic tests. The federal Health Insurance Portability and Accountability Act of 1996 prohibited group insurance plans from using genetic information to deny coverage, and in July President Clinton announced his intention to extend that assurance to individual policies. He stated, "It is wrong for insurance companies to use genetic information to deny coverage." Ricki Lewis, a freelance science writer based in Scotia, N.Y., is the author of several biology textbooks. She can be reached online at 76715.3517@compuserve.com.

Genetic Testing For Cancer Presents Complex Challenges

Interested in reading more?

Become a Member of