Reporting Genetic Results in Research Studies: Appendix A and B

Appendix A

Examples of genetic test results that should be reported to subjects:

Example 1: Mutations causing Marfan syndrome. Marfan syndrome is an autosomal dominant condition caused by mutations in the fibrillin gene FBN1 gene (chromosomal locus 15q21.1). The disorder is characterized by ophthalmologic and skeletal physical findings. Cardiac findings include a predisposition for aortic tear and rupture, which can be fatal. Ultrasound diagnosis of a dilated aortic root leads to monitoring and prophylactic surgery, which is life-saving.

Characteristics that make a Marfan syndrome genetic tests a good example of genetic information that should be shared with an interested subject are 1) That the penetrance approaches 100%; 2) Medical care can prevent death/complications, 3) the risk is well documented in multiple studies.

An additional feature that make this an important result to share is that the affected individual be may be undiagnosed, particularly among the 25% of cases that are estimated to be new mutations, without a family history to increase clinical suspicion of the diagnosis.

This example also illustrates some of the limitations of genetic testing. First, over 200 mutations in the FBN1 gene have been reported to cause Marfan syndrome or associated, milder phenotypes. In 2004 only 70-90% of subjects with Marfan syndrome have a positive test (Marfan mutation identified). A negative test (i.e. no Marfan mutation detected) therefore does not rule out the diagnosis.

More detailed information can be found at. http://www.geneclinics.org/

Example 2: Individuals homozygous for Factor V Leiden deficiency or heterozygotes who have an additional propensity to thrombophilia. The factor V Leiden mutation (F5G1691A) results in an identical amino acid substitution. This results in a slight increase in venous thrombosis in heterozygotes (RR 4-8), but a marked increase in the risk of venous thrombosis in homozygotes (RR ~80). Among heterozygotes, the presence of other inherited pro-coagulant tendencies such as protein C, protein S, and antithrombin deficiencies; prothrombin (PT) gene mutation, and elevated homocysteine increases the risk for thrombosis. Similarly, environmental factors such as estrogen exposure (use of oral contraceptives, hormone replacement therapy, and pregnancy) and surgery predispose to thrombosis. Known factor V Leiden allele carriers have their estrogen and surgery exposures carefully managed and may be treated longer for diagnosed thrombosis than other individuals.

Characteristics that make Factor V Leiden homozygotes a good example of genetic information that should be shared with an interested subject are: 1) dramatically increased risk of thrombosis; and 2) the ability to avoid or manage exposure to decrease morbidity associated with the thromboses, 3) this is a single mutation that can be accurately and cheaply tested by genetic techniques 4) the risk is well documented in multiple studies.

The question of notifying heterozygotes is less straightforward. Heterozygotes with additional known risk factors, such a PT mutations, should be given their genetic testing results. However, in the absence of other known pro-coagulant factors, the actual increased risk is small and must be weighed against the risk of hemorrhage from anticoagulant therapy. Further expert review should be considered to determine whether genetic test results should be disclosed for heterozygotes.

Examples of genetic test results that should not be reported to subjects:

• The results of genetic tests that demonstrate non-paternity should not be shared with family members. Although there is a conceptual advantage of identifying true fathers that have useful genetic information for the offspring affected, this does not outweigh the substantial psychological impact to families risked by sharing this information.
• An example of results that we do not support sharing with willing subjects is that of APOE. There are 3 common APOE alleles, 2, 3, and 4, and the 6 resulting genotypes have varying risk of Alzheimer disease. The risk for the 3/4 heterozygotes (`25% of Caucasians) is 2-3-fold increased and for homozygotes (1-2% of the Caucasian population) is substantially increased, perhaps 30-fold. However, APOE gene testing is not used clinically, even in patients at increased risk for Alzheimer disease. This is due to 1) the fact the outcome for a particular genotype is not known, 2) the current lack of a preventive strategy, and 3) the anxiety generated by a positive test in the absence of a benefit, particularly given the generally poor public understanding of the meaning of a risk factor, vs. presence of a disease gene. The same reasons that keep APOE from clinical use should result in scientists NOT sharing this result with subjects. Additionally, care providers of research study participants may be poorly prepared to interpret these results.

Examples of genetic tests which should be further considered to determine recommendations for release of results to:

• Factor V Leiden heterozygotes, as discussed above.
• Hemochromatosis homozygotes.
• Alpha 1 antitrypsin deficiency homozygotes
• Cystic fibrosis (CFTR) mutation carriers (heterozygotes).

Examples 1-2 indicate disorders where only a small percentage of individuals with the genotype will become symptomatic. Example 4 has a risk of having an affected child in mating with another carrier.

Appendix B

Genetic diseases for which clinical testing is available in more than two U.S. laboratories -- Sorted alphabetically. (Heart, Lung, Blood and Sleep disorders are in bold)