Even when a test can detect a mutation capable of causing a single-gene disease, the test may not be able to predict with certainty whether disease symptoms will appear or when they will appear. In such cases of incomplete penetrance, the specific mutation may be a necessary but not a sufficient condition for the disease to become manifest; other conditions must be present as well. These may be mutations at other gene loci or environmental factors; some or all of these other factors may be unknown. As a result, genetic test information on predispositions to a disorder has a potential for falsely labeling persons as being at risk for the disorder.
Many factors may modify the severity of single-gene diseases: the disease can have variable expressivity. The degree of severity (or extent of expressivity) cannot usually be predicted by a genetic test, even a test for a specific mutation; however, some specific mutations tend to be associated with specific levels of severity of the disorder.
Aside from problems that arise as a result of allelic diversity, incomplete penetrance, and variable expressivity, tests for genes implicated in single-gene disorders have other limitations. The accuracy of linkage tests will be impaired when an incorrect diagnosis is made in a relative or when the social father is not, in fact, the biological father. Linkage studies often have some uncertainty attached to them because the DNA marker linked to a disease gene can be separated by recombination. Such recombination events become more frequent as the physical distance between the marker and the disease gene increases. Tests will also be imperfect predictors if the laboratories performing them do not do so correctly and accurately.
Problems of penetrance and expressivity become even greater in testing for complex disorders in which multiple factors, of which the gene being tested is only one, contribute to the causation of the disease. In some of these disorders, a gene at a different locus (or genes at more than one locus) than the one being tested could contribute to causation. This is a form of genetic heterogeneity.
For some disorders, treatments of proven effectiveness will be available, but these treatments could be harmful to people who do not have the genetic disease (such as those in which a test was falsely positive). For many diseases, however, no definitive therapy will be possible when testing becomes available, although identification of the underlying genetic defect is likely to accelerate the discovery of future treatments. The duration of the lag period between testing and treatment capabilities will vary from disease to disease. For some disorders, the lag time will be so long that (1) individuals confronted with having to decide whether to be tested will not have any prospect of effective treatment to benefit them, and (2) they could not postpone having children long enough to see whether a treatment to benefit a prenatally diagnosed fetus will be discovered. Under such circumstances, nonmedical benefits and harms, as well as ethical considerations, dominate the decision about whether testing should be undertaken, both for individuals and for society.