Unapproved tests on a chip

    Prenatal genetic tests can now check for abnormalities in developing fetuses, but there is still no regulatory framework for them in the United States.

    For more than a decade, genetic tests have been on the way that would tell patients about genetic variations that might increase their susceptibility to disease. And all the while, bioethicists have been warning the US government of the impending need to regulate such tests when they arrive. Now hundreds of these tests are available, including powerful prenatal tests that use microarrays to scan hundreds of genes in unborn children. Yet despite the warnings, the tests remain effectively unregulated.

    Last December, Affymetrix of Santa Clara, California, became the first company to get approval from the US Food and Drug Administration (FDA) for a microarray chip for use as such a diagnostic device. The chip can check for genetic differences that might predict a patient's tolerance to drug treatment. Now, researchers at Baylor College of Medicine in Houston, Texas, are offering pregnant women a genome-scanning test that checks for abnormalities in developing fetuses (see page 733). But because of the way this test was developed, it is not currently subject to FDA review — and neither are tests made by private companies that can do prenatal screening but haven't yet been used to do so. The FDA has declined to say whether it intends to review these sorts of tests.

    One major point of confusion is whether the FDA has the power to regulate tests that are developed within a particular laboratory and are then used only in that lab, without being sold to outsiders. In 1997, the agency said that it would regulate these ‘home brew’ tests under the same rules that it uses to regulate other medical devices. But in 2003, when it issued further guidance on the ingredients that are used to make home brews, it decided that it didn't have the power to regulate them after all. Instead, the FDA said it would regulate only the ‘analyte-specific reagents’ — the ingredients, such as antibodies or nucleotide sequences, that make up the test. And it will only regulate the analyte-specific reagents if they are sold to the maker of the test; if the reagents are developed in-house, neither the reagents nor the test are subject to regulation by the FDA.

    “The unfortunate end result is a free-for-all in the marketing of genetic tests to the public, as well as widespread confusion among researchers and laboratories.”

    Companies that sell microarrays for others to use in genetic tests do have to register with the FDA, and must submit information supporting their ability to make the arrays properly to the regulator for approval. These rules also hold for companies or labs that make key components of genetic tests, such as gene probes. However, private labs that make and use their own microarrays for genetic testing are not subject to FDA scrutiny.

    Additionally, some companies are stretching the interpretation of the rules by marketing genetic tests that may be of dubious value (see Nature 426, 107; 2003). Because the companies make and use their own testing materials, they don't have to register under the regulations governing companies that sell microarrays to others. The unfortunate end result is a free-for-all in the marketing of genetic tests to the public, as well as widespread confusion among researchers and laboratories about their regulatory obligations.

    The introduction of prenatal microarray tests into this picture creates additional urgency for the regulators to act. If misdiagnoses occur, children may be born with an unexpected disease, or fetuses may be terminated on the basis of false information. Let's hope that such episodes don't have to be documented before the FDA acts to clarify its role as a regulator of genetic testing. If the agency finds that it lacks the authority to properly supervise the makers and users of home-brew genetic tests, then Congress should intervene to grant it that authority.

    Rights and permissions

    Reprints and Permissions

    About this article

    Cite this article

    Unapproved tests on a chip. Nature 438, 711 (2005). https://doi.org/10.1038/438711a

    Download citation

    Further reading


    By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.