Only a couple of years ago, people were amazed at the idea of genetically testing a fetus using no more than a pin-prick of the mother's blood. Suddenly, the tests have become big business: Illumina recently paid almost half a billion dollars to acquire Verinata, one of at least four companies selling fetal tests, with Sequenom, Ariosa and Natera. Perhaps as a sign of the money to be made, lawsuits have been flying among them. The talk is now of the tests' scope being determined not by the technology but by the market for it.
Non-invasive prenatal genetic testing was theorized at least 15 years ago, and demonstrated in principle by 2008. Sequenom announced [pdf] that it expected to launch a product in 2009, but that turned out to be so over-optimistic that it triggered an SEC investigation; the launch came in late 2011. Meanwhile, the concept was discussed in an AAAS article by Hank Greely and Jaime King in 2010. Greely's 2011 article in Nature, "Get ready for the flood of fetal gene screening" caused a ripple of concern, as did one shortly thereafter in Science Progress by CGS's Marcy Darnovsky with the prescient title:
One Step Closer to Designer Babies:
New Noninvasive Prenatal Genetic Testing Could Change Human Pregnancy Forever
Not enough people paid enough attention, it seems, until much more recently. In September 2012, New Scientist published an explanatory piece by Harriet Washington titled, "Do You Really Want To Know Your Baby's Genetics?" (an excellent introduction). In December, Wired ran a feature on the subject and just last month, two notable articles appeared, both with warnings in their titles. Scientific American said that "Fetal Genome Screening Could Prove Tragic" while MIT Technology Review introduced their survey as "A Brave New World of Prenatal DNA Sequencing."
What seems to have jarred at least some science journalists out of a kind of complacency is the clear demonstration that the technology is not just for autosomal disorders any more. (See George Estreich's posts, 1, 2, 3, for much more on the whole issue of testing for Down syndrome, the most publicized of them.) And choosing traits is getting much closer.
Initially, the tests on the market were specifically aimed at conditions in which the fetus has a duplicated chromosome other than the X or Y chromosome. Then some of them expanded to include several X- and Y-chromosome conditions. And now scientists from Tufts and Verinata have demonstrated that they can identify "fetal sub chromosome abnormalities" down to as little as 100 kilobases of DNA. That is not a product yet, but Verinata's chief science officer was quoted by Bloomberg as saying that they would give "serious consideration" to developing one:
"It won't be tomorrow," he said. "The study shows that this is no longer a technical issue for us. Now it's more of a question of how does the market want to drive forward with this."
Rival company Sequenom, which was the first on the market with an autosome test, seems even more committed to the process. Chief medical officer Alan Bombard told MIT Technology Review:
"We are just getting started with [abnormal numbers of five specific chromosomes], but we will be expanding this to single gene disorders."
It was only in December last year that The American College of Obstetricians and Gynecologists Committee on Genetics and The Society for Maternal-Fetal Medicine Publications Committee issued their guidelines on noninvasive prenatal testing, which acknowledged its potential but cautioned:
Cell free fetal DNA testing should be an informed patient choice after pretest counseling and should not be part of routine prenatal laboratory assessment. Cell free fetal DNA testing should not be offered to low-risk women or women with multiple gestations because it has not been sufficiently evaluated in these groups. A negative cell free fetal DNA test result does not ensure an unaffected pregnancy. A patient with a positive test result should be referred for genetic counseling and should be offered invasive prenatal diagnosis for confirmation of test results.
Solid, conservative, sensible. And much too late. That recommendation came after Sequenom had already sold some 60,000 tests and was on pace to double that number this year alone. The companies like to cite the statement as approval; Sequenom even includes the phrase "should not be part of routine prenatal laboratory assessment," though careless readers might overlook it, because they recast the sentence to imply that informed patients should choose to have the test.
This train seems to be careening down the tracks, and the medical establishment is just not keeping up with it. The NIH has launched a five-year, $25-million program to study the partly related issue of genomic screening in newborns, which is welcome, but already may be too late. In five years will screening be common (if not actually routine) not for newborns but for first-trimester fetuses? And how are we, as a society, going to react to that?