By Matt Jones
NEW YORK (GenomeWeb News) – Since premarket studies are not likely to fully answer questions about the safety and effectiveness of genetic tests, those that are sold directly to consumers should be regulated through a system that incorporates risk-stratification and post-market surveillance, according to a new policy opinion proposal published today in Science.
Direct-to-consumer genetic tests pose unique problems for potential regulation, and effective oversight will require flexibility and cooperation from government to account for the variations and complexities of the tests, the paper's authors argue.
As the authors point out, barriers to oversight by the US Food and Drug Administration include the lack of available data to support the sort of premarket approval FDA offers for new drugs, as well as an inability to control the way the tests are used and will be used in the future.
Despite these potential difficulties, the government has made moves in the past year that signal some type of regulation is on the way.
FDA has sent letters to a number of genetic testing companies saying that their tests are medical devices and require regulation, and the Government Accountability Office conducted an investigation which appeared to show certain DTC companies providing misleading or deceptive claims about their tests.
There is a need for data about the kinds of effects genetic tests could have on consumers and physicians, but trying to generate that data through the premarket model could cause a range of problems and burden industry, the authors — including lead author Amy McGuire, associate director for research at Baylor College of Medicine's Center for Medical Ethics and Health Policy — explained.
"No one single simple regulatory strategy is going to work here… because the nature of these tests is that they're testing for many different things in one test kit," McGuire told GenomeWeb Daily News today.
"Delaying consumer access to DTC tests until robust premarket data are available would inhibit product development and could effectively regulate DTC companies out of existence," the authors argued.
They argue that premarket studies cannot answer many questions about the safety and effectiveness of genetic tests, so a risk-based approach that focuses on post-market surveillance is the best way to approach such tests.
Their proposal is that FDA's premarket review would focus on higher-risk tests, and that less risky tests would receive less premarket scrutiny but would still be covered by other oversight mechanisms, such as a national registry.
Under this strategy, premarket assessment would focus on identifying tests with the potential for egregious harm – those with uncertain validity or utility that could alter the course of medical treatment – and keeping them off the market until more information is known about them.
All of these tests should be analytically valid, accurate, and reliable, and clinical claims made by firms about their offerings should be substantiated, the authors argue.
"The real concern we have is that in order to get effectiveness data on a lot of these genetic variants is going to require quite a lot of time and there are intermediate measures that can be taken to ensure that they are sufficiently safe to be on the market," said McGuire. "Then, if you're aggressively tracking them and having surveillance and collecting post-market data you can grant an agency like the FDA the authority to respond if data amasses that there is a lack of safety or a lack of effectiveness."
Exactly how the risk analysis model at FDA would function for genetic tests is not entirely clear.
But what McGuire said the authors do not want is to have a regulatory situation where regulators have to "look at every single SNP that is being tested for and make a case-by-case determination.
"I think that would be overly burdensome for the agency. They don't have the staffing or expertise necessary to do that," McGuire continued.
That is where the risk stratification approach will pay off, they say. The classification system would probably include high-risk, moderate-risk, and low-risk categories. Some tests would require higher regulations, and others would require only limited premarket regulation or post-market scrutiny.
McGuire said it is a real concern that if a regulatory system is too stringent, if the premarket requirements for drugs were applied to genetic tests, for example, regulations could deter companies from offering DTC genetic tests at all.
McGuire added that regulation is not necessarily viewed as "the big bad enemy," and that many DTC companies themselves "would welcome some form of regulation, and that the devil is in the details about what that regulation would look like."
She also said that because so many of these firms sell their tests over the internet and across international borders, "federal regulation would be a great starting point," but "ultimately some international harmonization on regulations" would be ideal for consumer genetic tests.