To understand this debate, one needs some background on the history of carrier testing in the United States, a history with two strikingly different testing experiences, both of which are influencing plans for CF testing.
The first experience dates to the early 1970s, when about a dozen states enacted poorly conceived laws that mandated testing blacks for sickle cell trait. The testing programs usually targeted black school children or black couples seeking marriage licenses.
These laws paid little, if any, attention to the importance of providing genetic counseling services, to protecting the privacy of the test results, to the problem of the inadvertent discovery of nonpaternity, and to other sensitive issues. As a result, they drew sharp criticism, which, in turn, helped to stimulate enactment of a federal law that offered matching funds for sickle cell testing programs to those states that embraced voluntarism, respected privacy, and provided counseling.
The second experience also can be traced to the early 1970s, when Michael Kaback, then at Johns Hopkins University, and other geneticists pioneered community-based screening programs to identify Ashkenazic Jews who carried the gene for Tay-Sachs disease (an autosomal recessive neurodegenerative disease of childhood). These programs, by contrast, won high praise. Today, about 25,000 Ashkenazic Jews in the U.S. are tested each year, and the number of children born with this devastating disease is about one-tenth that of the pretesting era, due to the high testing rate in the at-risk population and selective termination of affected fetuses.
Unfortunately, testing for CF is not as straightforward as testing for either sickle cell trait or Tay-Sachs disease, at least from a clinical standpoint. While a single mutation (a deletion of the three base pairs resulting in the loss of a phenylalanine residue at position 508) accounts for approximately 70 percent of CF carriers, an unknown number (currently more than 60) of different mutations constitute the other 30 percent. This poses a problem for testing, which currently can identify carriers of the 508 mutation and a few other mutations, but which cannot identify carriers of most of the less common CF mutations.
Assuming a 70 percent carrier detection rate, CF testing then presents two problems. First, the test would detect only 49 percent (.70 x .70) of at-risk couples, and second, the couples consisting of a known carrier and a person who does not carry one of the screened mutations arguably face an added anxiety in their pregnancy. (At 70 percent detection, the risk to a white couple of northern European extraction is about one in 300, which compares with a pretest risk of one in 2,500.)
The technical limits implicit in a test that could identify only 70 percent of carriers was the primary reason that leaders of the American Society of Human Genetics (ASHG) issued an October 1989 policy statement advising that the test be offered only to persons with a family history of CF. Mindful of malpractice litigation, the statement pointedly declared that routine CF carrier testing of pregnant women and other individuals was "not yet the standard care in medical practice" (T. Caskey, et al., American Journal of Human Genetics, 46:393, 1990). It called for pilot programs to study laboratory, educational, and counseling aspects of screening.
Realizing that as the ability to detect CF mutations improved, pressures to provide testing for this common, serious genetic disorder would mount, the National Institutes of Health sponsored a workshop on population screening for the CF gene in March 1990. Noting that the testing could identify only 70 to 75 percent of carriers, the group recommended against population-based screening for individuals with a negative family history.
The workshop participants also expressed concern that there are substantial limitations on the ability to educate people regarding an "imperfect test." Nevertheless, the group concluded that such screening and counseling "difficulties would be substantially reduced if testing could detect at least 90 to 95 percent of carriers." It also declared that pilot testing programs were urgently needed (New England Journal of Medicine, 323:70-71, 1990).
During 1990, hopes for quickly developing a test that identified 95 percent of carriers faded, but substantial progress has been made. For example, by last autumn, medical geneticist Arthur Beaudet at Baylor College of Medicine was using a set of DNA probes to identify 84 percent of carriers, and it now appears likely that a 90 percent detection rate is a realistic near-term goal.
When this goal of a 90 percent detection rate is reached, the test would identify 81 percent of at-risk couples and miss only 1 percent. It would also place those couples consisting of a known carrier and a partner who has tested negatively at a relatively low risk (about one in 1,000) of bearing a CF child.
As the detection rate moves inexorably to the 90 percent level, I think the argument in favor of offering the test to persons regardless of a family history outweighs that against. At this point only two of the NIH workshop participants' four reasons against wide-scale testing would remain: that the frequency of the gene varies substantially across ethnic and racial groups (which complicates the utility of the test and the counseling process), and that there will be problems in educating physicians and counseling patients about test results.
The first problem is a relatively straightforward one to solve. It requires gene frequency data, which should emerge rapidly. Already, a consortium of about 80 laboratories around the world has collected much of these data.
As for the second, the debate will focus on whether physicians, especially obstetricians, and other health-care providers are able to appropriately offer the test, which includes helping patients understand the results. In essence, this is a debate about the future of genetic counseling--a relatively new profession--and the manner in which genetic testing is offered in our society.
As the array of carrier tests grows and the cost of testing drops, genetic counseling will earn a place in primary care. The argument that primary care providers will not be able to provide adequate genetic counseling must be measured against the patient's right to know about a potentially helpful test.
At some point (which may ultimately be decided by the courts) this right to know invokes a corresponding obligation on the part of the physician to inform (and counsel or refer for counseling) couples about the test. It is at this juncture that the duty to inform outweighs concerns about miscommunication in the effort to inform.
In summary, the debate about when, to whom, and how to offer CF carrier testing merits full discussion. CF testing is a paradigm for many other genetic tests that will someday be routinely offered as part of preventive medicine. This is why pilot testing and counseling projects are needed now. The consensus that emerges from the discussion of CF testing will be a benchmark for other efforts. Of course it is important to train physicians, nurses, and other health- care providers to counsel patients about the implications of being a CF carrier, but as the detection power of the test grows, the patient's right to be informed becomes paramount. We are rapidly approaching that point.
Medical geneticist and attorney Philip R. Reilly is executive director of the Shriver Center for Mental Retardation. Located in Waltham, Mass., the center conducts basic research in mental retardation and provides services to mentally retarded individuals.