The New Prenatal Testing: No Risk of Miscarriage

Amniocentesis [2], in which a needle is thrust into a woman’s womb in order to test for fetal abnormalities, can be nerve-wracking. Although the risk of miscarriage is small, especially at high-end facilities [3], the discomfort it evokes in pregnant woman is sometimes primal. “My womb is a sheltered place, a protected place, the one place my baby should be safe and undisturbed. Yet here was this doctor with this thing, this needle, shattering it,” recalls a woman quoted by anthropologist Rayna Rapp in her classic study of amnio in America [4].

Another option, chorionic villus sampling [5], can be done earlier in a pregnancy than amnio (between 10 and 12 weeks rather than typically after 15 weeks) but also involves a needle through the abdomen, or a thin tube through the cervix, to remove a small piece of the placenta. And so until now women who wanted definitive information on Down syndrome and other chromosomal abnormalities had no choice but to undergo an invasive procedure.

But that is likely to change. New prenatal tests are on the way that rely on a blood sample taken from the mother’s arm. They should be commercially available within the next few years. In the long run, as the number of conditions they can detect expands, the new tests may shunt the old invasive ones to the side. Why take a big needle to the belly if a little needle in the arm tells you everything you need to know about your fetus? But if that leads to more prenatal testing overall, is it a good thing?

The new tests take advantage of an alluring discovery: From early in pregnancy, a small amount of DNA from the fetus enters the mother’s blood and circulates in her body. This represents a source of genetic material for examination that scientists didn’t know about until relatively recently. The first genetic testing using this DNA is likely to focus on a small group of problems, including Down syndrome. But over time, testing could look for the full range of chromosomal abnormalities detectable with amnio or chorionic villus sampling [5]. All of this may relieve women of one source of prenatal anxiety—though it won’t make getting the results of genetic tests any less fraught.

Amniocentesis dates from the late 19th century, when physicians in Germany first used it to relieve dangerous levels of pressure in the womb by withdrawing excess amniotic fluid. By the mid-20th century, amnio was also viewed as a window into fetal health. Doctors in the United States and elsewhere used it to assess fetal lung development and to determine the gender of fetuses at high risk for hemophilia (the chances of a boy carrying the disease are much greater). They also saw an opportunity to look for Down syndrome, once the chromosomal cause was identified. Rapp explains that advocates began a drive to offer the test to women at a higher-risk for bearing Down syndrome children: In a pivotal wrongful birth case [6], a woman whose baby turned out to have the disease successfully sued her doctor.

Since at least the 1990s, amnio rates have been falling, from more than 130,000 procedures in 1990 to fewer than 40,000 in 2005, according to the American College of Obstetricians and Gynecologists. (These figures do not include data from a number of states.) That’s partly because more soon-to-be parents are relying on early, non-invasive screening tests [7]. They can begin in the first trimester, and don’t involve anything more invasive than ultrasounds and blood tests. But though they predict fetal risk for selected abnormalities including Down syndrome, they can’t offer certainty.

The real genetic testing winner would be a trifecta: definitive, early, and non-invasive. And all of that may soon come on the market. In one approach, researchers take blood samples from pregnant women, then amplify the free-floating DNA molecules, looking to see if any chromosomes are overrepresented. (In Down syndrome, the fetus has an extra copy of chromosome 21.) “We let each molecule vote for a chromosome and we look essentially for voter fraud,” says Stephen Quake of Stanford, who developed the technology with graduate student Christina Fan.

In research [8] published in 2008, Quake’s team successfully picked up in a small sample all cases of Down syndrome, and two other major disorders called trisomy 18 [9], and trisomy 13 [10]. The testing took place during the second trimester, but Quake says it could in theory be used in the first trimester, too—“almost as soon as Mom knows she’s pregnant.” Stanford has since licensed the technology to two companies, Artemis Health and Fluidigm. (Another company, Sequenom, which appeared to be far along with its own cell-free, fetal DNA approach, has since admitted [11] to “mishandling of R&D test data and results” and is now reevaluating its test.)

Prenatal testing is barreling ahead on another front, as well: an approach called array testing, which dramatically boosts the information load available to parents-to-be. Array testing currently relies on fetal DNA retrieved from amnio or chorionic villus sampling [5]. Laboratories place the DNA on a chip, and can quickly suss out whether the fetus has any of hundreds or thousands of genetic disorders. Many of these, like Angelman [12], Prader-Willi [13] and Williams syndrome [14], are rare but can cause severe impairment. Those diseases, and many others, would be missed by traditional genetic analysis following amnio or CVS, the method that preceded array testing, because that relies on a profile of the chromosomes called a karyotype.

The National Institutes of Health is funding a large, multi-center trial to compare array testing to karyotype analysis. It could end up validating the higher-tech approach. And already a few pioneers, including Baylor College of Medicine in Texas and a Washington-based company called Signature Genomics, offer array testing to women who’ve had abnormal sonograms or who otherwise want more information. The downside at this point is the price tag—about $1,600 a pop from Baylor and $1,850 from Signature, which many insurance companies don’t cover.

Someday, could scientists combine the two new technologies? That is, perform advanced array testing on genetic material retrieved by way of a mother’s arm? It won’t be easy: The free-floating fetal DNA in the mother’s blood may not be pristine enough for array tests without further purification. And whole fetal cells, while present in maternal blood, are hard to come by, says Sau Wai Cheung of Baylor. Still, the Baylor group hopes to marry arrays to non-invasive blood tests in five to 10 years.

If that happens (and the odds seem good), it could dramatically alter the calculus for prenatal testing by offering much more information available earlier and with no added risk of miscarriage.

Of course, reams of data can be a mixed blessing. Array testing can pick up anomalies that no one knows with certainty how to interpret. Particular abnormalities on chromosome 16, for instance, have been linked to autism [15], but may also be found in healthy people. Several geneticists told me that they advise against testing for it. DNA testing in general can crystallize anxieties by dangling both too much information and nowhere near enough. That’s not going to change based on the size of the needle or the science behind the test.