Materni21, Harmony, Verifi, and Panorama: Myths and Realities

Some relatively new tests promise to find Down syndrome and other chromosomal conditions very early in pregnancy.  They are called Noninvasive Prenatal Screens (NIPS), and have been hailed as the “holy grail” of prenatal tests. Just weeks after a fetus’s heartbeat is audible by doppler, the mother can have her blood drawn to look at something called cell-free fetal DNA. In a now-billion dollar industry, Sequenom (Materni21), Ariosa (Harmony), Natera (Panorama) and Verifi all offer NIPS, promising unprecedented accuracy, better information, and general reassurance. The website for the Materni21 test boasts that the product offers “No confusion. Just simple, clear results.”

A 2012 article in the Atlantic began with this explanation:

“Prenatal diagnosis — the ability to diagnose abnormalities before a baby is born — is undergoing a revolution due to the recent arrival of tests that can accurately detect fetal genetic abnormalities, including Down syndrome, by testing the mother’s blood.”

When I became pregnant again–already having a child with Trisomy 21–I seriously considered this new technology, wondering what it would mean for me to have another baby with a genetic difference.  It could mean quite a bit or very little, but at a minimum, getting a definitive diagnosis might change my prenatal care decisions.

When I researched these tests, however, I was shocked at what I found.  The idea that NIPS are an amazingly accurate diagnostic tool is simply not true.  The reality, I found, is a very troubling mix of public and medical misunderstanding, which may well be rooted, at bottom, in pharmaceutical companies looking to maximize profit by mainstreaming their products.

Myth #1: NIPS is great because it is noninvasive way to get a diagnosis.

Reality: NIPS is called NIPS because it is just that: a screen.  (Read the ACMG guidelines here.)  The number of false positives and negatives does not make it a reliable diagnostic tool.  The current ACOG guidelines recommend that NIPS be followed up with a diagnostic test, either chorionic villus sampling (CVS), or amniocentesis. Both of those tests are invasive and carry a risk of miscarriage. In reality, there is no foolproof way to know whether your baby has a trisomy. Amniocentesis and CVS are very, very reliable. Still, mistakes happen.  Bottom line: NIPS can only give you a statistical likelihood that your unborn child carries a limited list of genetic conditions.

Myth #2: NIPS is amazingly accurate.

Reality: NIPS seem to be good at ruling out the existence of a genetic condition, but “positive” tests are more complicated.  NIPS accuracy rates really refer to the tests’ statistical sensitivity, which is not what most of us think of when hearing the word “accuracy”. An individual’s chance of truly having a condition after a “positive” result (also called the Positive Predictive Value, or PPV) depends on the incidence in the population at large.

Here is an imaginary example.  If a test is 95% accurate at predicting redheads prenatally, how does one know the actual chance of being a readhead after testing positive?  Not 95%.  One has to know how many redheads there are in general.  Say 10% of population of country X has red hair, and we test 200 mothers there.  Then we know that 20 fetuses truly have red hair.  A 95% accurate test will find 19 of those redheaded babies, and miss one.  It will also incorrectly identify 5% of the remaining 180 as having red hair when they actually do not—9 babies.  The positive test result group is then made out of 19 truly redheaded babies and 9 not redheaded babies.  Any individual baby’s real chance after a positive result (or PPV) of having red hair is really 19/28, or 67%.

Do this again with brunette babies in the same country, with an incidence rate of 40%—80 babies with brunette hair.  76 will test positive, 4 missed.  Of the remaining 120 babies, 6 will be incorrectly identified.  The chance of a baby who tested positive as a brunette is then 76/82, or 92%.  When the incidence rate in the general population goes up, the PPV goes up.

Now, say that one in 1,500 babies in country X have purple hair.  Do the math for a 95% accurate test, and you get a PPV of something like 1.25%.  That means a mother who tests “positive” for her baby to have purple hair, in this scenario, actually only has a 1.25% chance of delivering a baby with purple hair.  When the incidence rate is already low, the PPV is lower.

Same test accuracy rate in each scenario, but very different PPV, because the reality of a “positive” test result depends on how often a condition actually occurs in a given population.  (I know, I’m sorry your head hurts, mine did too.)

What does this mean?  Unless a mother is already at high likelihood, looking for conditions like aneuploidy (like Down syndrome) or Turner’s syndrome is a lot like trying to find babies with purple hair.  NIPS tells us what we already know: Most women are not carrying a baby affected by these genetic conditions.  When you’re looking at a condition that only 1 in thousands have, the PPV of NIPS is quite a bit less impressive than the test companies would have us believe.  Depending on a woman’s age, a “positive” NIPS result could mean the chances of actually having a baby with Down syndrome could be 50% less. Mark Leach at his blog Down Syndrome Prenatal Testing offers a very good explanation of the statistics in this post.  Here is another more general article aimed at explaining the confusing statistics behind medical tests and screens.

Myth #3: My doctor recommends it and understands NIPS better than I do.

Reality: It turns out that doctors, just like the rest of us, are actually not very good at understanding the nuances of statistics.  In one study, just 14 out of 61 doctors, hospital staff and medical students could correctly answer a hypothetical statistics question.  Certainly, it doesn’t appear that the companies offering NIPS tests are running out to disabuse confused medical professionals of the idea that their tests are not as “accurate” as they seem.  Do your own research.

Myth #4: NIPS will analyze the baby’s DNA.

Reality: Actually, the test runs on small fragments of placental DNA in the mother’s bloodstream. In most situations, the fetus and placenta share identical DNA. However rare, placental mosaicism does occur. In that case, a woman could get an incorrect NIPS result. Her placenta may carry a trisomy while the actual baby does not, or vice versa.

Myth #5: NIPS can tell you if you are going to have a healthy baby.

Reality: NIPS can tell you absolutely nothing about the health of your baby. Trisomy 21, for example, results in medical issues for some, and absolutely none for others. In addition, NIPS tells us nothing about the myriad other conditions that a fetus can carry but remain undetected until birth or after.

Myth #6: My test result came back “positive” so it must be correct.

Reality: Materni21 terms its results “positive”, “negative”, or “additional” findings, but in reality, these words appear to be more than a branding tool than medical terminology, akin to how Starbucks calls its smallest drink a “tall” to give the impression of value. Because there is no such thing as a “positive” NIPS result, the use of the word “positive” is actually fairly misleading. See Myth #2.

Fundamentally, I do think that these tests stir up some very serious ethical issues about disability, as well as how our society views pregnancy and birth. Plainly, this technology begs the question: Who gets welcomed into our society, and who must undergo “screening” before even drawing a single breath?  Yet, this is the reality of all prenatal testing.

My concern is the power of perception to recreate reality.  If these companies are allowed to perpetuate these myths with impunity, then I fear they will become the foundation for actual practice.  Perhaps my fear is already reality.  How many women have chosen abortion after a NIPS result, who would not have if they had truly understood the statistics?  How many women misunderstood these statistics and go through needless anxiety and worry because they’ve had to prepare for a reality that never came to pass?  NIPS is very good at “reassuring” mothers that their babies are genetically typical, but does that come at a cost?  With every “negative” test, how much stigma compounds against those who actually live with those genetic differences?

Even within the disability community, we seem to have bought the pharmaceutical companies’ advertising ploys, hook, line and sinker. I worry that by perpetuating the companies’ marketing ploys, we are inadvertently creating more room for ethical lapse.  The tests are here for good—there is no closing that box, and I see no point in arguing that.  We might as well help families get accurate information and strive to point out the inherent anti-disability bias so prevalent in today’s prenatal care practices.

Is NIPS an advancement?  For a woman who wants to test for those specific conditions, I do think that NIPS has value over traditional screenings.  Is NIPS revolutionary?  Only if we pretend it is.

***************************************

ETA 9/22/2015:

This is my most popular blog post, by far. I want to ask that everyone to have the most sensitivity and grace as possible when commenting. It is often difficult for those who have had true positives to hear the sense of rejoicing that comes with false positive results. Praying for health or a false negative, while completely understandable, does not make the correlate true: having a baby with health concerns or a positive diagnosis does NOT mean that your prayers have gone unanswered. I want the comment section here to be a safe place and to assure families getting diagnosis that life will still be beautiful and worthwhile and that support is out there. Concepts such as “health” and “perfection” are socially influenced and variable; there is no one right way to come into this world. xo

***************************************

If you have had a negative NIPS experience, or feel that NIPS was incorrectly or misleadingly explained by the company’s marketing material, you can report it here.  Prenatal tests fall under the “medical device” category.

Here is a very clear article from Katie Stoll, genetic counselor, explaining the specific statistical issues behind NIPS.

Lettercase offers up-to-date information and materials on prenatal screening and testing for families and medical providers.

Mark Leach is a bioethicist who writes frequently about NIPS on his blog.

If you have found this post after getting a “positive” result for Down syndrome in particular, you can go to the Down Syndrome Diagnosis Network (DSDN) for accurate information and support.

If you have found this post after a Harmony, Verifi, or Materni21 test and are confused, feel free to comment or email.  I will try to send you to the appropriate resources.