by Keith Syverson
In March 2010, the National Institutes of Health (NIH) announced that it is creating a voluntary national Genetic Testing Registry to provide easy public access to information regarding the availability and utility of genetic tests. The registry, which is slated to be available in 2011, will collect information from clinical, commercial, and research testing organizations and display it in a database geared towards physicians and patients. The announcement came almost two years after the Secretary Advisory Committee on Genetics, Health, and Society (SACGHS) published recommendations suggesting that the NIH establish a web-based registry for laboratory genetic tests. Ideally the registry will include information about the analytical validity, clinical validity, and the clinical utility of genetic tests. This information is particularly important for consumers who wish to utilize direct-to-consumer (DTC) genetic testing options such as 23andMe or Navigenics. Although the database is a step in the right direction for ensuring that patients and doctors receive accurate information regarding genetic testing, the registry may not alleviate the problems associated with direct-to-consumer genetic testing.
As currently conceived, the registry allows testing providers to voluntarily submit information regarding the individual tests they offer. They can submit as little or as much information as they chose, but the registry "encourages submitters to provide explicit molecular information about the test they perform and to cite published support for their assertions." However, no data from individuals' genetic testing will be collected or displayed. Each test will be given a unique number which will allow for cross-referencing with scientific publications and electronic health records. The content and quality of the data reported will not be reviewed by the NIH and the presence of a test in the registry will not constitute an NIH endorsement of the test's validity. Outside of these basic guidelines, no details have been made available regarding what a report would look like or how much information a testing provider can (or will) provide.
For a genetic test to benefit a patient, the test must be analytically valid, clinically valid, and possesses clinical utility. Analytical validity refers to the "accuracy with which a given laboratory rest identifies a particular genetic variant." More generally, a test is analytically valid if it can consistently distinguish between genetic variants (for example, between an adenosine nucleotide and a cytosine nucleotide). To be analytically valid, a test need only prove that a specific letter of the genetic code is located at a certain location in the genome. The test is clinically valid if it shows that a genetic variant correlates with a specific trait or disease. Lastly, clinical utility refers to the likelihood that using the test results will lead to a beneficial outcome.
Some commentators argue that without proper regulation and validation, DTC genetic tests can undermine public confidence in those clinical genetic tests that are widely accepted by doctors such as tests for cystic fibrosis or Huntington disease. DTC tests are often rushed to market without proper validation and as a result, many may lack clinical utility. Because, even if the tests are analytically valid, a test can be clinically invalid if the underlying correlations between a genetic variant and a condition or trait are based on flawed studies. Common problems with studies connecting genetic mutations with particular traits include: small sample size, poor study design, and incompatible follow up analyses. A study with a small sample size can miss minor contributions of single or multiple mutations. Small sample sizes can also produce imprecise or incorrect estimates of the magnitude of the observed effects. Poor study design result in the failure to include proper controls which can add unaccounted for variables such as environmental exposure and differences among populations. Moreover, follow up studies that analyze different variants produce results that can neither confirm nor deny the original analyses. For example, an association between a genetic variant and schizophrenia was identified in people of Irish descent and reportedly confirmed in independent European studies. The confirming studies, however, reported different risk alleles, which makes comparison difficult.
Without regulation or enforceable guidelines as to what constitutes a reportable genotype-phenotype association, DTC genetic testing companies are free to offer tests and provide reports for whatever disease or trait they choose. This has led some commentators to argue that the incentive for companies to profit from genetic associations drives the premature marketing of genetic tests based on inconclusive data. For example, the three major DTC genome-scan companies 23andMe, Navigenics, and deCODEme all have different standards for what constitutes a reportable association.
In additions to problems with the clinical validity of the tests, difficulties can also arise for the consumer evaluating the clinical utility of a test. With traditional clinical tests, the clinical utility of a test or procedure is evaluated by physicians when they decide whether or not to order a test. If a test does not produce useful results, generally a physician will not waste the patient's time or money ordering a test. However, DTC genetic testing bypasses this checkpoint, requiring the patient or consumer to make judgments on the clinical utility of a particular test based on information provided by the company offering the test. This poses a danger when, as is the case of many traits or diseases tested by DTC genetic testing companies, the clinical validity of a test is uncertain.
Ideally, the Genetic Testing Registry will collect information regarding the analytical and clinically validity of a genetic test and provide consumers with up to date accurate information. Unfortunately, genetic testing companies are not required to participate in or submit any information to the database and the NIH will not review the submitted information for content. As a result, it's questionable how much information will be submitted and doubtful whether any of the problems associated with DTC genetic testing will be remedied. However, at least one company is on board with the registry: 23andMe co-founder Anne Wojcicki stated that "We welcome the news of the Genetic Testing Registry . . .. 23andMe has always been committed to providing individuals with the information they need to make the most of their own genetic information. We look forward to working with the NIH on this project."
If other genetic testing companies are as enthusiastic about participating in the project as 23andMe, there is the potential for effective self-regulation among testing companies. Moreover, because the database will also provide links to ordering information, there is a significant incentive for companies to participate in the database. As one commentator points out, the collection of information in one place will help standardize test offerings and spur competition between companies to provide reliable accurate information about the tests they provide. Similarity, if consumers rely on the database, "inclusion in the [registry] may well become a de facto requirement for any commercial genetic test provider, even if it is not converted into a legal requirement."
The registry has the potential to facilitate self-regulation between testing providers, but there is still a significant chance that a person can misinterpret the results of a test without the aid of a doctor. Although the registry may have accurate information regarding testing, it cannot replace a discussion between a physician and a patient about what a particular result means and how he or she should respond to it.