The costs, benefits, and ethical issues of increasingly sophisticated prenatal genetic tests.
hen I became pregnant with my now 3-month-old son, I was a walking Wikipedia of the biology of conception and gestation. I could measure the passage of time with my ovulation cycle. I could flowchart the process of in vitro fertilization that had led to the conception of our son. I could even apply my biology degree to choosing a model combination of natural and medicalized care. Yet I still went through my pregnancy knowing virtually nothing about prenatal genetic testing.
When I first encountered a genetic test, I thought it was another blood draw for cholesterol or glucose or even an STD. My ignorance was blissful. I was 37 years old and 16 weeks pregnant. My doctor sent me off to the lab with an order for something called a quad test. It was only when the results came back that I realized it was a test for genetic conditions—specifically, for Down syndrome (trisomy 21), Edwards syndrome (trisomy 18) and two types of neural tube defects called spina bifida and anencephaly. The results were normal for three out of the four conditions. At the time I didn’t even know enough to feel relieved. All I saw, highlighted in obnoxious fluorescent yellow, was a set of numbers: 1 in 26.
The numbers were followed by a sentence that stated the fetus I was carrying had a 1 in 26 chance of having Down syndrome. Blissful ignorance turned into disturbed confusion. What exactly did these numbers mean? Was this very high risk or medium high risk or what? A Google search and a scan of baby boards revealed women whose numbers were much higher (and whose probabilities were therefore lower)—1 in 135, 1 in 57, 1 in 90—sharing their anxiety as they waited for the results of more definitive tests like amniocentesis or chorionic villus sampling (CVS). According to those boards, my 1 in 26 result was a near foregone conclusion that my baby would be born with Down’s.
My doctor recommended I proceed with an amniocentesis. I had to find out on my own (back to the baby boards) that while the quad test is merely a screening test that indicates risk for Down’s, an amniocentesis can definitively diagnose or rule out Down syndrome and a host of other conditions. My husband and I ruminated on the numbers. If we flipped the 1 in 26, then our baby-to-be had a 25 in 26 chance of not having Down’s. That certainly sounded more positive. Was this enough of a risk to warrant an invasive procedure like amnio? More important, did we really need to know if the baby would be born with Down’s?
It wasn’t enough and we didn’t need to know. The odds were still in our favor that we would have a healthy baby, and we were prepared, or willing to get prepared, to raise a child with Down’s. We declined the amnio and I went back to trying to enjoy my pregnancy.
It’s Just a Blood Test
I was 28 weeks pregnant and living in a new city when the genetic testing question once again reared its troublesome little head.
“I see your quad test came back positive for Down syndrome,” my doctor said. She was getting up to speed on my records.
“It did. At first we were worried but then decided the odds are still good that our baby is healthy,” I reached out to hand her my birth plan, ready to move on to other issues.
“Have you heard about the cell-free DNA test?”
“No. What’s that?”
“It’s a new test, the latest thing—non-invasive and doesn’t carry the miscarriage risk of amniocentesis.” Her eyes were as bright as her big smile. “I believe it’s something like 99 percent accurate.”
“Wow, that’s good to know. But I don’t think I want to do any more tests.” The walking Wikipedia of biology in me was intrigued, but purely from an intellectual standpoint.
“Well, so you know, it’s just a blood test. It might help for peace of mind, or to help you prepare if your baby does indeed have Down’s.”
I looked at my husband. He blinked back at me. It sounded so easy. It was like she was recommending the free coffee in the waiting room. Have some. It’s simple. Okay, thanks. Why not?
“Okay, thanks. Why not?”
“Exactly. Why not? I can order the test right now.”
“Oh. Well, maybe we’ll take a couple days to think about it.”
“Of course. But I’ll go ahead and order it. You can go straight to the lab if you decide to have it.”
The path of least resistance was to do the test, because why not? We had chosen a hospital that prioritized natural childbirth and intervention only when necessary, and we assumed this meant they’d have a critical view of elective testing, especially genetic testing. Not so. I later realized that how this went down was an insult to informed consent. This illustrates how commonplace such optional genetic tests have become in even “natural” hospital maternity settings. It also illustrates the many issues raised by the availability of new genetic tests like the cell-free DNA screen, issues that are unfortunately being glossed over in both popular discourse and doctor-patient applications of the tests.
The Evolution of Prenatal Genetic Testing
Genetic testing in pregnancy has been around for almost half a century, since genetic amniocentesis first became available in the 1970s. Prior to this, amniocentesis had been used for non-genetic purposes including testing something called Rh compatibility. Amniocentesis was the first widely-applied method for genetic testing, made possible by the ability to culture fetal cells from amniotic fluid and then generate a picture, or “karyotype,” of certain chromosomes.
This was followed by chronic villus sampling (CVS) in the 1980s. CVS is a procedure with similar diagnostic power to amniocentesis, but rather than drawing from the amniotic fluid, CVS uses cells from the outermost membrane of the amniotic sac. CVS’ main advantage over amniocentesis is that it can be done a bit earlier in pregnancy (between 8 to 10 weeks as opposed to 14 to 20 weeks).
The next major advance came in the 1990s with the advent of blood serum genetic tests like the quad test. By measuring levels of fetal and placental proteins present in the mother’s bloodstream, these serum tests offered a non-invasive first screening step before amniocentesis. There are a handful of serum tests offered many different ways, including the First Trimester Combined Screen, which is a blood test plus a special ultrasound scan, the second trimester Quad test, which is the one I had at week 16, and a combination of the two called Integrated Prenatal Screening. The tests are currently standard for pregnant women 35 years and older, and are routinely offered to pregnant women under 35. But the variety of tests can be very confusing, and they have a limited predictive success rate of 80 to 90 percent.
Enter the cell-free DNA (cfDNA) test. The test was first developed in 2012 and is just now becoming more widely available. Also a blood test, the cell-free DNA test differs from the aforementioned tests in both method and accuracy. Instead of measuring protein levels, this test is much more direct in that it analyzes fetal chromosomes released into the mother’s bloodstream from disintegrated placental cells. (Note that the DNA comes from placental cells, not fetal cells, as was originally claimed by the companies who developed the test.)
This makes the cfDNA test a significant scientific innovation. The scientist who discovered fetal chromosomes floating in maternal blood likened the discovery to finding out that a car’s engine is not just under the hood. It’s also a potential leap forward for pregnant women. The test is much more accurate than previous serum tests, predicting 99 percent of all Down syndrome (trisomy 21) pregnancies, 98 percent of all Edwards syndrome (trisomy 18) pregnancies, and 65 percent of all Patau syndrome (trisomy 13) pregnancies. Incidentally, it can also tell the sex of a fetus long before genitals can be discerned through an ultrasound. In the cases of the often-fatal Edwards and Patau syndromes, early detection through cfDNA followed by amniocentesis can give women the option of a first-trimester abortion, an option which could spare the pain of birthing a stillborn baby or a baby likely to die within the first month or year of life. The cfDNA test can be given as early as ten weeks of pregnancy, potentially offering a more streamlined, accurate, and early indication of risk for these and other genetic conditions.
So What’s the Catch?
I happened to mention to my friend Emily, an internal medicine doctor to whom I tell everything, that we decided to do the cfDNA test. “Our doctor made it sound so simple,” I explained.
“OK.” Pause from her end. “You’ve done your own research on it, right? As long as you know and feel alright with the false positive rate and the pros and cons.”
Silence from my end. I hadn’t done any of my own research. I decided to correct that, and what I found was not exactly pretty.
My doctor had told me the test was “something like 99 percent accurate.” This is true. What she didn’t tell me was that even with this high level of accuracy, the cell-free DNA test is not a diagnostic test. The gold standard of diagnosis is still the amniocentesis, which is nearly 100 percent accurate. What this means is that if I took the cfDNA test and it came back positive I would still need to have an amniocentesis to confirm the results. And if the test came back negative, there would still be a .1 to 1 percent chance of that being wrong as well. Basically I would be right back where I started with a few more degrees of certainty, but I would still have to face an invasive test for confirmation.
Apparently, my doctor was not really to blame for misleading me on this front. A slew of recent news articles have falsely hailed cfDNA as a potential replacement for amniocentesis. This implies not only that cfDNA is as accurate as amnio, which it isn’t, but also that it tests for the same things, which it doesn’t. CfDNA tests only for trisomies 21, 13, and 18, certain sex chromosome abnormalities, and the gender of the fetus, while amniocentesis tests for all of those, a wider range of chromosomal abnormalities, and neural tube defects.
My research led me to another important fact: The test is not covered by insurance. Had I blindly followed my doctor’s advice I would have received a fat bill for $500 or possibly much more, since out-of-pocket costs for the cfDNA test can be as high as $2,000. CfDNA is simply too new to be underwritten by insurance companies. And while the Affordable Care Act mandates at least partial coverage for prenatal testing, including genetic testing, it may be a while before cfDNA is included in this mix.
Billion Dollar Business
By this point my blood was simmering. What I found out next brought it to a boil. The cfDNA test is getting a lot of attention right now because results of a major clinical trial were recently published in the New England Journal of Medicine. The study provides the strongest evidence yet that the cfDNA test is more accurate than standard blood tests in predicting Down’s and Edwards syndrome for both low- and high-risk pregnancies. (Previous studies had focused on women over 35 whose pregnancies were considered high-risk.) Essentially, the study lays the groundwork for wide adoption of the cfDNA screen as a priority genetic testing option in prenatal care.
This would all be positive if it weren’t for a serious conflict of interest. Rarely mentioned in all the hype about the study is the fact that the trial was fully funded by Illumina, one of the laboratories that developed the test and the “800 pound gorilla of sequencing companies.” What’s more, eight of the 13 study authors, including the lead investigator, are paid by Illumina as employees, consultants, or advisory board members. So much for independent research.
Even if the science itself is sound (one of the editors of the New England Journal of Medicine described the study as “well done”), what’s clearly suspect is the way the results are being distorted in the press: “New Prenatal Test Edging Out Amnio” and “No More Needles in the Stomach” are two examples. These overstatements are no accident. The companies that produce the test have been aggressively marketing it, including pitching to reporters and direct marketing to consumers through YouTube, as well as other social media. The money these companies are putting into large-scale clinical trials and aggressive marketing seems to be a wise investment; Forbes estimates the market for DNA-based prenatal tests is currently $1 billion, with the potential to grow to $6 or even $8 billion.
Were my doctors getting kickbacks to grow the market for these tests? I didn’t really believe this to be true, but given what I had learned the thought did cross my mind. In any case, someone was going to get rich off these tests and it clearly wasn’t going to be me.
Keeping Up with the Genomes
Though I felt angry and had, I also felt validated by what I had found. My experience was not just an example of one eager-beaver doctor and a clueless patient trouncing on the value of informed consent. There was a systemic problem afoot.
Simply stated, when it comes to prenatal genetic testing, scientific innovation, along with corporate marketing, are getting far, far ahead of good clinical practice. Four Stanford researchers explained, “Unlike various European countries… the USA has very limited direct regulation of either reproductive technologies or genetic testing and no national health care system. As a result, large professional societies, private medical insurers, and for-profit companies largely govern the uptake and integration of new technologies in prenatal practice.” The result is what I experienced—haphazard and irresponsible implementation of the cfDNA test at week 28, as well as no information about options before I took the quad test at week 16.
The fact that my doctors didn’t have clear guidelines to follow is unacceptable. Protocols for the most responsible ways to offer the cfDNA test and all genetic tests are supposedly being developed by the American Congress of Obstetricians and Gynecologists and other professional societies. They must be developed, and soon—ideally in collaboration with women’s health advocates, bioethicists, and disability rights advocates.
From the patient’s perspective, I didn’t have appropriate resources to help me understand what I was being offered. I was never referred to a genetic counselor. And though I took an eight-week childbirth course and skimmed What to Expect When You’re Expecting, I never learned what to expect from genetic testing. There’s a simple solution to this: prenatal health classes: two- to three-hour classes that include an overview of good care and that share objective information about available genetic tests. These would be an important addition to the option of genetic counseling because they could offer a holistic view of how genetic testing fits into overall prenatal care. Prenatal health classes should be standard offerings at hospitals and clinics across the country, much like childbirth and lactation classes have become.
Protocols and education are going to be especially important as more and more DNA-based genetic tests become available. Laboratories are already trying to sequence the full fetal genome early in pregnancy. This means that eventually a cfDNA test could tell you not just whether a fetus is at increased risk for Down’s, but also what hereditary diseases he or she might develop later in life.
It’s worth remembering that 95 percent of all pregnancies in this country are healthy births. This high percentage is partly due to the fact that many fetuses with aneuploidy (more or less than the normal two copies of each chromosome) are spontaneously miscarried before the second trimester of pregnancy. So we have to ask ourselves whether we really want to know if a fetus has a genetic condition so early in a pregnancy. And eventually, will we really want to know if a baby-to-be has, say, an 80 percent chance of developing Huntington’s disease before the age of 50? Exactly how much information is too much?
Information from genetic tests is already raising expectations about how much we can control pregnancy outcomes. In the past two years there have been at least three high-profile “wrongful birth suits”—cases in which families who received negative results from genetic tests have successfully sued for millions because their babies were born with genetic conditions. Does this mean we are headed toward a society where people with disabilities are considered “wrongful” and will suffer even more discrimination and receive even less systemic support than they do now? And what about women who choose to forego genetic tests or who willingly choose to have babies with genetic conditions? Will this be the newest way that women can be condemned for being bad mothers? On another front, what do we make of the fact that genetic testing is currently more accessible to some low-income women on Medicaid than to women with private insurance? Are there some subtle eugenics at play?
These ethical questions used to be theoretical. They’re not anymore. Standard protocols, education, and informed consent won’t necessarily help us with this terrain, but a robust national conversation will. At stake is no less than the kind of society we want to live in. We may want a society that guarantees a woman’s right to choose on all fronts, including our right to terminate both unwanted pregnancies and wanted pregnancies with debilitating health conditions. I know I do. We may also want to live in a society that guarantees disability rights for both physical and intellectual disabilities, where resources are prioritized to allow people to “live a full life with a range of talents, capacities, and difficulties.” I know I do. Will we be able to have it both ways?
In the end my son was born healthy. We beat the 1 in 26 odds. We’ll never know what the cfDNA test would have shown, but we now understand its potential and its limitations. Would I do the test if we got pregnant again? Maybe. By then there may be even more tests to wrap our heads around. By then may there be more balanced dialogue to help us understand them.
I love the thought that went into this and the amount of research you did, considering I have a child with Down syndrome (total surprise, no prenatal testing) and his younger sister (I did the cfDNA testing). I was surprised to learn the fetal vs placental differentiation, since I, too, was led to believe it was fetal.
However, I really wish you had used the correct diagnosis of Trisomy 21 or Down syndrome instead of the colloquial “Down’s,” which is incorrect.
Larli thank you so much for your comment and the correction – I was wondering about this since as I was doing research I saw it listed as both Down’s and Down Syndrome – I figured one was more correct than the other but didn’t know which so I really appreciate you letting me know.
Yes, the cfDNA test relies on fetal chromosomal DNA released from placental cells specifically, which is significant because this can lead to false positives from something called “placental mosaicism” in which the placenta can actually be shedding DNA with three copies of certain chromosomes even though the fetus itself may not have aneuploidy in those chromosomes.
Here’s a good explanation of placental mosaicism: http://www.downsyndromeprenataltesting.com/when-cell-free-fetal-dna-isnt/ and an abstract of a scientific article that touches on it: http://www.ncbi.nlm.nih.gov/pubmed/24667696
We did IVf as well, and did do genetic testing at the embryo level (I’m secure enough to tell you that we did this in great part due to the fact that we were doing a surrogacy and the issues surrounding those are quite different for a variety of reasons). They give you a report and that’s it. No say otherwise….I have my own issues knowing that one of those embryos had trisomy 21. We had five “normal” and five “abnormal”. You are so well-spoken (and written).
Thanks for your sharing your experience Amy – embryo level testing is another area entirely and I’m glad you brought it up! I suspect embryo testing will become even more available for people who choose surrogacy and IVF. It makes a lot of sense especially given the added “investment” of resources, time, emotions etc. But wouldn’t it be nice to get more counseling and advice along the way instead of just receiving a report!
This is a super interesting and informative read! Thanks for getting the conversation started about an extremely important and relevant topic. Not having any kids yet, I wasn’t aware of all of the genetic testing options available so this has opened my eyes to yet another issue that clearly needs more research and more informed docs to help foster better patient-provider conversations.
Hi Alyssa thanks for reading and commenting! Hopefully when your turn comes you’ll have access to a good genetic counselor if that’s what you’ll want, and who knows, maybe even the pre-natal health care classes I mention will catch on and become a real offering.
Thank you for sharing your narrative. You really should think about writing about your experience for health care providers. The medical community is moving toward a notion of patient centered care/shared decision making and a key question that we discuss is how much information is too much information and what sorts of information ought to be share with patients, etc. I think your story shows that the answer to this question is likely different for all patients, but that patients are perfectly capable of understanding the uncertainty of modern medicine and digesting complex data –if its made available to them.
Emily thanks so much for this comment (and — full disclosure — for being a key impetus of this piece!). What you’re saying about moving towards patient centered care/shared decision making is really interesting, and I have to confess my first reaction was “why is this an innovation, shouldn’t this be the default?” and then, understanding the history of medical care in this country, my second reaction was “that is so great to hear and how can I learn more.” My biased opinion is that patients in general want more information rather than less, and it’s easier to choose to ignore unwanted information that is offered by care providers than it is to find out wanted information.that is not offered by providers. Also speaking for myself I may have been capable of understanding complex data and uncertainties, but I would really have appreciated professional help in doing so!
Great article! There’s often concern about industry funding university research but no one talks about the amount of “research” private industry is publishing in the world. Some of which are glorified ads. I’m fortunate my husband has a PhD in Biology and we have access to other doctors and experts and it was still challenging to deal with the emotional struggles of what do all these statistics mean for you. I once asked my OB, what was it like before all these tests? She asked an older peer, and he said you just waited. Given the roller coaster of infertility journeys there were times I wished for ignorant bliss.
Hi M Bautista, I have to confess I was scandalized when I found out that Illumina funded and powered the study about cfDNA in the New England Journal of Medicine. I guess I was still in the blissfully ignorant state of thinking that this was the exception rather than a growing trend (is it?)
Really enjoyed reading your article. I live in Mississippi, where health care is generally okay, but not as good as other places in the country. I’m 33 weeks pregnant with my second child. With both pregnancies, I was not consulted on whether I wanted to have genetic testing done. The doctor just ordered blood work and then told me “everything looked good” after the tests came back. I feel like I really have to make an effort to ask lots of questions. Also, the only classes available in my city for pregnant women/couples is an epidural class. No childbirth classes or lactation classes. Definitely no prenatal health classes. I would hope that this will change over time.
Alicia congratulations on your pregnancy! And thanks for the important reality check that childbirth and lactation classes are not actually widely available outside certain states and urban centers. I, too would hope that this will change over time and would like to get involved in any advocacy efforts to help make that happen. The fact that the only class available in your city for pregnant people is an epidural class is so telling about how far we still have to go to win health care that treats women like whole people. Thanks for sharing your story and I hope you enjoy the rest of your pregnancy.
Great review of childbirth in the modern world, seemingly a “naturally” confusing process. What brave parents you are, and thank you for providing valuable information on the ever evolvong world of costly and not necesarry procedures/tests
Your research did not lead you far enough, I’m afraid. Many insurance companies, including Medicaid and United Healthcare do cover the free-cell DNA test for those over 35 or with a history of abnormalities. For women, like me, who have experienced labor ing and delivering a stillborn child, this test is a relief. “My research led me to another important fact: The test is not covered by insurance. Had I blindly followed my doctor’s advice I would have received a fat bill for $500 or possibly much more, since out-of-pocket costs for the cfDNA test can be as high as $2,000.”
y I appreciate your comment. I can’t imagine the pain and grief of laboring and delivering a stillborn child and I know it sounds trite to say in a comment section, but I am sorry for your loss.
It’s an important correction that some insurance companies cover the test, though I’m not sure it’s accurate to say “many” as I know for a fact that the test is only covered by Medicaid in certain states.
It was my hope in writing this article to expose the complications behind the test, not to dismiss it outright. I do think it’s an important advance for pregnant families, but I believe it must be applied responsibly and correctly by caregivers and in the context of proper counseling.
As I wrote: “In the cases of the often-fatal Edwards and Patau syndromes, early detection through cfDNA followed by amniocentesis can give women the option of a first-trimester abortion, an option which could spare the pain of birthing a stillborn baby or a baby likely to die within the first month or year of life. The cfDNA test can be given as early as ten weeks of pregnancy, potentially offering a more streamlined, accurate, and early indication of risk for these and other genetic conditions.”
Thanks for all the research and the summary. My wife and I are worried sick over the sheet of paper containing our cfDNA results the show a positive for downs. Our doc does believe and state that the test is almost a certain indicator. This, after the 12 week ultrasound and NT measurements turned out to be normal. Her midwife also said that she’s seen a 1-1 correlation between a positive cfDNA test, the amniocentesis test and an actual occurrence of downs in the baby. This link seems to disagree with that assertion (http://www.downsyndromeprenataltesting.com/adam-wolfbergs-research-says-dont-offer-new-testing-for-down-syndrome-to-all-moms/) but despite my basic understanding of how publications work I’m very confused.
We were patted on the back and sent out the door with a referral to a genetic counsellor / amniocentesis clinic without being able to see or talk to our doctor. I guess we have to sweat it out by ourselves while waiting for our next appointment without being able to talk to anyone about it that can understand and spare us the fake concern. Those of our family that live outside the US are incredibly surprised that these tests are even offered to low risk individuals (my wife is 29 and we’re both really healthy) who presented a normal 12 week ultrasound. Their comments aren’t helping wipe away my darling wife’s tears.