Dr. Michael Watson
Although the terms are often used interchangeably, testing and screening have two distinct uses. Genetic testing involves the analysis of symptomatic individuals or those with a family history, whereas screening uses a population-based approach. Ethical issues associated with both of these forms of testing include test validation and test performance. The intended use of the test should be specific; it might therefore be helpful to articulate the intended use as if FDA approval were required.
There are 2 stages through which a genetic test can be translated into clinical practice.
Stage 1: includes population-based research to establish scientific links between genes and diseases. Laboratories would not have to be approved under the Clinical Laboratory Improvement Amendments of 1988 (CLIA) because they do not report to the families, but Institutional Review Boards should oversee patient involvement. The FDA regulates manufactured devices and drugs, and provides guidance for informed consent requirements, but genetic "home brew" tests often fall through the regulatory gap. Moreover, while CLIA regulates how laboratories practice, it does not address the issue of clinical validity.
Stage 2: involves clinical investigation to establish the analytical and clinical performance characteristics of the tests. This stage usually requires working with patients and families in order to correlate clinical risks or presentation with genetic and non-genetic components. It often requires that results regarding the scientific meaning of the test be provided at the time of disclosure. These issues become more complicated when genetics labs interface with specialty practices where there may be less communication.
Another aspect to consider is that our experience is often based on classical Mendelian diseases. Because many Mendelian diseases are rare, requirements (what kind?) are often "loosened" to allow lengthening of the investigative stage, and statistical power of validation studies are impacted. There is also a lack of financial incentive for CLIA labs to offer tests for rare diseases, which complicates the transfer from the research to the clinical laboratory environment.
Determination of clinical utility is conducted during this stage of test development; clinical validity and utility issues are crucial, because with multiple variations or mutations, there can be a variety of clinical phenotypes. This may require the study of large populations. Ultimately, the determination of clinical utility requires the correlation of clinical information with test information, and this could be problematic in cases where someone else owns the database of information.
Stage 3: involves the development of practice standards for tests that are found to be adequate and of utility at Stage 2. This involves the development of guidelines for testing and who should be tested.
Parkin testing is a good case study in policy development. In order to develop appropriate policies, large collaborative studies, data collection methods, and ascertainment of long term susceptibility outcomes are needed, as parkin testing involves complex mutations and phenotypes. (A similar example is the cystic fibrosis gene which is a "complex gene.' Parkin mutations may be even more difficult to test for than cystic fibrosis mutations since there is extensive variability in the mutations in parkin.) Ethnicity is also a factor because it is not uncommon that mutations will vary among racial groups.
In conclusion, Dr. Watson offered suggestions for how to develop policies for testing. In order to obtain the necessary power in clinical trials of therapeutics, researchers will need large collaborative multidisciplinary efforts to increase subject recruitment. There is a move toward more national collaborative studies, however, to date, there has been little incentive to study susceptibility genes because of the long term commitment required; currently only the NIH has the resources to invest in long term outcome studies. Testing of those individuals not at risk in the general population should also be done.
As one measure to ensure safe and effective testing, the American College of Medical Genetics (ACMG) has tried to keep abreast of the issues by standardizing tests at early stages, driving research agendas, and developing practice guidelines when the community has enough knowledge to apply a particular test. The need for education of all involved, from neurologists to the lay public is tantamount.
Dr. Kimberly Quaid
A genetic test is any test that reveals genetic information and may be performed on DNA, RNA, or protein, or any substance that indirectly reflects gene function. Genetic information includes both DNA sequence information and inferences that can be made from knowledge of the sequence. This can reveal the clinical condition (clinically apparent or latent) of individuals, families, close blood relatives in both preceding and succeeding generations, or in groups that share a common ancestry. It can impact on how an individual relates to other members of their family, and can require an individual to balance the loss of personal privacy against benefit to a relative, especially in linkage cases.
Genetic tests are also uniquely personal in that they can impact individual perceptions of health, body and worth, an influence that can extend over the lifetime. The information can also be used as a means of identification. The ability to perform these tests can create the need to make decisions or choices about education, the future (such as whether or not to have children, choosing among reproductive techniques, or testing for future illnesses) or what to do for a living.
Dr. Quaid outlined the types of genetic testing as:
Presymptomatic testing is performed on a person who has no symptoms of a specific disorder at the time of testing. The test determines whether or not he or she has a mutant gene, with the goal of identifying their risk for developing a disease in the future. It involves looking for genetic mutations that have a high penetrance (usually autosomal dominant conditions). These tests are usually highly sensitive and specific with few false negatives or false positives. Presymptomatic tests will usually provide a clear family history. Ideally they should help individuals in planning for the future, allow them to take advantage of whatever preventive strategies are available, and help to make reproductive choices.
Presymptomatic testing has now enabled us to identify healthy individuals who are very likely (virtually 100% as in with Huntington's Disease) to develop a devastating and debilitating disease at some point in the future for which we currently have no treatment or cure. Ethical issues are thus inherent in presymptomatic testing; are we better off knowing our fate? This is a very individual question. The issue of respect for personal autonomy involves everyone's right not to know, the reluctance to test children, and issues of informed consent. The question of psychological costs to those tested is also an important ethical issue as well as that of prenatal testing for late onset disorders, e.g. Huntington's or late onset Alzheimer's diseases as examples.
Huntington's disease was mapped to a previously unknown genetic location on chromosome 4 in 1983. Following this discovery, testing was only offered through an NIH-funded research protocol designed to measure the psychological, social and economic impact of testing. Ten years later in 1993, testing guidelines were published by the Huntington Disease Society of America and the World Federation of Neurology. The development of guidelines took was multidisciplinary, involving neurologists, psychologists/psychiatrist, geneticists and genetic counselors, and the protocol included a physical exam, pre-test counseling, and informed consent. Results were communicated in person, and follow-up assistance was made available, such as support groups or mental health professionals.
In this example, the role of the counselor was to provide relevant genetic information, taking into account the individual's current level of knowledge and the novelty and complexity of the genetic information. The counselor also helped people to understand the consequences to their lives and families, and provided emotional support and practical help with decision-making. In early-onset Alzheimer's disease, testing was first offered through a few centers with extensive experience in testing for Huntington's disease and followed similar protocols.
In this case, the individual has no symptoms of a specific disorder at the time of testing, but does so to determine the presence or absence of a mutant gene. In contrast to presymptomatic testing, if a mutant gene is present, the person has a higher probability, rather than a certainty, of developing symptoms of the disorder at some time in the future provided they live long enough. Susceptibility testing is more relevant to a much larger at-risk population and yields more ambiguous information. Cautions against prematurely testing for susceptibility occur with limitations in sensitivity and specificity coupled with a lack of treatment options.
In conclusion, Dr. Quaid noted the following ethical concerns: the general lack of knowledge of the public about genetics, the lack of oversight, the direct marketing of genetic tests to consumers, and the fear of discrimination. The introduction of widespread presymptomatic and susceptibility testing in neurogenetics should be preceded by research examining the ethical legal and social issues related to the introduction of such testing.
Dr. Andrew Singleton
The goal of the lab is to find the genes that contribute to Parkinson's disease. Early onset disease is more familial than late onset however PD does seems to be a very complex genetic disease, potentially with many genes interacting to cause the disease. He presented all the PARK gene loci identified thus far, primarily through family linkage studies.
Dr. Singleton then discussed how mutations are found, stressing that not all mutations have a direct effect. One of the ways to test if a mutation is having a real effect is to look at the protein it encodes and its potential involvement in PD. It is still very unclear what variations contribute to Parkinson's disease. In detecting mutations there are several techniques that are easy to do and a laboratory does not need large amounts of equipment. Dye-terminator sequencing for example, examines genes for small sequence alterations. The methodology is easy to perform - the data are reliable and are easily assayed. Other methodologies such as gene dosage (copy number assays) methodologies are currently a lot more difficult to perform and data are often variable. Though this method is being used more and more in the laboratory it often results in many variations on assays and has no gold standard.
There are four common gene dose methodologies which he discussed in further detail:
Array-based technologies are used for sequencing and dosing. Several companies are now developing custom re-sequencing chips. Sequence and dosage variations can be assessed simultaneously, the method is quick to perform however the reliability is still unclear at this point and it is prohibitively expensive. The ease of use does make this methodology very promising.
Test interpretation is of central concern, and Parkin is an illustrative example. Parkin contains most mutations originally implicated in juvenile Parkinson's disease. The original report from Matsumine, et al (1997) was based on linkage data using multiple families, and was backed up by a discovery of a parkinsonism family with a genetic microdeletion in this region. This was followed closely by the discovery of the gene by Kitada et al (1998) showing that it was recessive and one needed to have two copies. In general, when a gene is linked to PD the area in the genome is shown to harbor a mutation by linkage and the mutation is identified. These steps are typically performed in the laboratory with a lot of experience. The burden of proof in this step is very high. Once the mutation is identified follow up studies are performed with more mutation screening. As a result one could find that the mutation shows unequivocal segregation with the disease, one could find that the mutation shows an unclear segregation with the disease, or, finally the mutation was found in a single case and not identified in controls by specific assays. The burden of proof here is lower however there is a rapid need to confirm positive results.
Over the next several years there was an increase of publications investigating the parkin gene mutation with varying degrees of proof. The conclusions were, however that one copy was confirmed with some risk associated. Parkin mutations were most common in Parkinson's patients and you also see heterozygous single mutations in the general population . A similar story is emerging for the DJ-1 gene which leads one to wonder if the same will be true for other recessive loci. Reduced penetrance of other linkages is likely to complicate the story further when more genes are discovered (PARK10, PARK8).
Dr. Singleton offered the following important facts to consider:
In conclusion, medium-throughput screening and dosage sequencing is becoming more reliable and reproducible but challenges still exist with data interpretation, suggesting that the pathogenicity of each new mutation should be assessed on a case-by-case basis.
Dr. William Seltzer
Athena's mission is to improve the quality of the lives of patients with neurological disorders by providing the best possible diagnostic services to their physicians (they do not market directly to patients). Athena has offered neurogenetic testing since 1989 and currently provides over 100 testing services for diseases ranging from movement disorders to metabolic myopathies. They are a fully mandated and voluntarily accredited by CLIA, the College of American Pathologists (CAP) and the Joint Commission on Accreditation of Healthcare Organizations (JCAHO) clinical laboratory with stringent commercial standards, and have frequently been cited for excellence. They have designed blinded studies to look at the analytic parameters if they are able to detect mutations. Once they have performed the blinded studies they train technicians to perform the tests. Their goal is to transfer from research and development to the clinical lab, and then to evaluate their tests through a series of blinded clinical evaluations.
The first Charcot Marie Tooth test launched in 1993 is an example of how a test helped improve diagnosis. There are now seven different tests offered by Athena. The first ataxia test was launched in 1994, and there are currently eleven different genes with tests available. Parkin alterations will provide the first significant genetic link to Parkinson's disease providing important access to discoveries. Other known or unknown loci may also prove to be important in the future.
The Parkin gene is well described (Kitada, et al) and there are nearly 200 publications reporting mutations found in multiple populations. Parkin testing will be a valuable tool to aid clinicians in diagnosing early onset Parkinson's disease in symptomatic individuals less than 45 years of age, at a rate of 7,500 cases per year (that's 15% of all Parkinson's ). The positive results of testing will definitively diagnose "parkin disease" and provide information on inheritance. As a widely available test, clinicians would have access to a rapid, reliable diagnostic tool.
Athena's Parkin test uses a two-step methodology: quantitative PCR analysis to detect duplications and deletions using both internal reference standards and controls; and gene sequencing to detect small sequence variants which uses both normal and positive controls. The test exceeds rigorous quality control and quality assurance standards as required by CLIA, CAP and JCAHO. Based on their experience thus far, the test could provide 300 patients per year with a diagnosis of parkin disease. Statistics from their test to date shows that the majority of parkin tests yield definitive results with 4% positives, 80% negatives, 7% carriers and 9% indeterminate. The reports are developed and reviewed between Athena scientific staff and experts in the field offering detailed information to assist clinicians in interpretation. The reports contain sections on interpretation, technical results, methods, comments, references and a glossary. Important adjunct services offered by Athena include consultation on test ordering and result interpretation from their in-house geneticists and genetic counselors. Athena offers expert-reviewed product information and sponsors grand rounds at academic institutions as well as web-based educational material.
Dr. Seltzer summarized that this was an exciting time in neurogenetics and Parkinson's disease, potentially mirroring successful test development for ataxia and CMT. It's important to have a qualified service provider who could partner with clinicians in diagnosing early onset PD, and Athena welcomes a continuing dialogue.
Mr. Jeffrey Martin, spoke to the group from a patient's perspective and as someone who received the test for the parkin mutation through NIH. Mr. Martin stressed that his is one person's perspective and he could not presume to speak for all PD patients; being provided with test results is a personal preference and participation in research does not require that the patient get the results.
For rare autosomal dominant cases, testing may help life planning in terms of family, insurance and other financial matters, or career choice. People don't necessarily want to know their genetic makeup; just because the technology is within our ability does not mean that its use will enhance our lives. Perhaps an important aspect of clinical trials is getting the genetic information and banking it to provide resources throughout the country.
The patient community needs to show leadership and caution. The use of the parkin test as an aid to clinical diagnosis is questionable, and imaging is a more reliable test parameter. The test is not necessary for determining diagnosis. Mr. Martin also discussed concern for the high cost -it won't change treatment, and the implications for children are not great because their risk is low. Other risks include implications for the insurance agencies - if we get information that predicts one is susceptible to diseases what affect will that have on our ability to get insurance and or employment?
There are benefits of knowing in terms of early diagnosis; this could also allow patients to pursue new and different neuroprotective strategies and could potentially affect treatment in the future.
Parkin mutation testing should not be recommended lightly, if testing is performed there should be physician training as to how to counsel patients, or direct patient counseling.
Ms. Elizabeth Thomson
This issue has been raised with other diseases; the question is whether or not to test the population once a mutation has been found for a particular gene. There are two ways that health policies can develop according to Wilfond and Nolan (1993).
The extemporaneous model of policy development is the more descriptive model and acknowledges the role of stakeholders, advocating their own interests. Phenylkotenuria (PKU) screening and maternal serum alpha-fetoprotein (MSAFP) testing are examples of extemporaneous models. PKU screening was implemented too early because the inventor and consumer groups went to each state to stress the importance and need for the test. It was thus implemented before we knew whether the test really worked. MSAFP testing was the same. It became the standard of care - not because of the evidence but a because of a legal risk analysis done by gynecology groups concerned that they may be sued by patients with children with neural tube damage.
The extemporaneous model stems from increased interest of professional, consumer and commercial groups as well as liability issues. The tests become "established" through increased utilization which, when combined with reimbursement, drives the standard of care.
The other model, the evidentiary model, is a more prescriptive model and refers to an explicit approach in policy decisions. Few standard of care models use an evidence based model. The goal here should be to try to encourage everyone to look at evidence to support the use of a specific test or not.
The evidentiary model is based on epidemiological and clinical studies with a clear evaluation of the underlying normative issues. Public and professional consensus drives the standard of care in this model which in turn drives the utilization and reimbursement.
The use of new genetic tests is often driven by the following:
The issues and questions then become:
Assumptions are made about the value of genetic test information and there are variable motivating factors for the use of genetic tests. There is a fair amount of predictive uncertainty and health professionals are not always aware of how to interpret test results.
Early gene discovery research is often conducted in "high risk" families or other potentially biased populations. This can yield higher prevalence/penetrance rates than is confirmed in wider populations. For example, the BRACA1/2 prevalence rates started at 85-90% but are now confirmed at closer to 27-55%. Also early studies are often conducted on Caucasians with extrapolation to other populations. Gene mutation rates can vary substantially, affecting sensitivity and specificity of tests for various populations. We need to know the critical facts, not just biological, but also ethical, clinical, psychological, social and economic. The data needs to be analyzed without bias and the facts need to be placed in the context of diverse social and cultural norms; professional and public consensus should be reached.
There are still many uncertainties as the range, prevalence and penetrance of mutations/variability is not known. The risk of stigmatization and discrimination of those who test positive but who may never become ill should be addressed.
Ms. Thomson made an overall call for research in PD testing beyond biology:
While the current test is targeted toward neurologists, PD patients are extremely well informed and sophisticated. The majority of patients know about the test so government, advocacy and investigators have an obligation to the community to weigh in on this issue. All presenters agreed that testing procedures are relatively straight forward compared to the challenges of interpreting data. Where do we go from here?
Dr. Seltzer agreed that information on who is ordering the test, their background, when did they first present, etc, would be important to collect. Athena is not set up to gather this information but they are willing to partner with of the community to get this information. A research facility could then analyze the data. Mr. Martin suggested it was essential to make sure patient reports show the test limitations of interpretations.
Dr. Shoulson noted the strong group consensus on the use of the test for research. Most of us agree that this kind of testing should take place in a research environment where we can gain invaluable information. It was suggested that companies with tests could companies help fund DNA banking efforts.
Some other relevant comments/needs:
Matsumine H, Saito M, Shimoda-Matsubayashi S, Tanaka H, Ishikawa A, Nakagawa-Hattori Y, Yokochi M, Kobayashi T, Igarashi S, Takano H, Sanpei K, Koike R, Mori H, Kondo T, Mizutani Y, Schaffer AA, Yamamura Y, Nakamura S, Kuzuhara S, Tsuji S, Mizuno Y. (1997). Localization of a gene for an autosomal recessive form of juvenile Parkinsonism to chromosome 6q25.2-27. Am J Hum Genet. Mar; 60(3): 588-96.
Kitada T, Asakawa S, Hattori N, Matsumine H, Yamamura Y, Minoshima S, Yokochi M, Mizuno Y, Shimizu N. (1998). Mutations in the parkin gene cause autosomal recessive juvenile parkinsonism. Nature. Apr 9; 392(6676): 605-8.
Lincoln SJ, Maraganore DM, Lesnick TG, Bounds R, de Andrade M, Bower JH, Hardy JA, Farrer MJ. (2003). Parkin variants in North American Parkinson's disease: cases and controls. Mov Disord. Nov;18(11): 1306-11.
Wilfond BS, Nolan K. (1993). National policy development for the clinical application of genetic diagnostic technologies. Lessons from cystic fibrosis. JAMA. Dec 22-29; 270(24): 2948-54.
Last updated March 1, 2011