A good friend of mine recently started freaking out, and asked for my advice. “Help!” she says to me. “I had the MaterniT21 test and it shows the baby has Trisomy 16! What do I do?”
Uhhh… I’m not an obstetrician, and had never heard of MaterniT21. But as a general non-believer in screening tests, I was immediate skeptical. Was there really a blood test that could absolutely tell you that a fetus is abnormal? Really?
MaterniT21 is a test that is actually now 3 years old, developed by Sequenom, a San Diego-based biotech company. Other similar tests that came out after Sequenom’s include Harmony by Ariosa, Panorama by Naterna, and verifi by Illuminata/Verinata. (Sounds like new brand-name perfumes or trendy hip-hop tunes.) The test is a DNA test, which is important to understand. It examines fragments of fetal cells found in the mother’s blood. The cell fragments are from the placenta, not from the fetus. Still, DNA information is more accurate than many ultrasound findings, many of which are termed “soft” markers for chromosomal abnormalities. The company website claims a 0.2-0.3% false positive rate for Down Syndrome and other abnormalities.
Sounds awesome! Sign me up! Here are a few reasons why I won’t (or wouldn’t, since I’m three and out) be getting the MaterniT21.
1. Many patients who are offered the test don’t understand statistics. I don’t really, either, but someone on babycenter.com does: (http://community.babycenter.com/post/a32275977/clarification_about_maternit21_positivefalse_positive_results)
False positive rate = false positives/all tests. False positive rate does not equal false positives/all positives
It’s counter-intuitive especially for low-incidence cases such as Down Syndrome (DS). For example, study shows that Maternit21 has 99.8% specificity, which means 0.2% false positive rate. It also has 99.1% sensitivity, which means it can detect 99.1% of all the DS cases.
Since launch there have been about 3500 tests. I don’t know what’s the incidence of DS in the tested population but since the test is only recommended to high risk group which usually have a 1/2 to 1/200ish odds of DS, let’s assume the allover DS incidence being 1/100 for the tested population. So by definition, if the study statistics can translate into real life situation, so far the test would have identified:
# of True positve cases = 3500 x (1/100) x 99.1% = 35
# of False positive cases = 3500 x 0.2% = 7
Bottom line is: If you got a positive result from MaterniT21, there is still a 7/(7+35)=1/6 chance that it’s wrong. And that is why a positive result needs confirmation by amnio/cvs.
2. Testing freaks parents out. Mothers almost always undergo prenatal testing, usually in the form of a triple screen or quad screen, which tests for the possibility of the most common birth defects. This blood test, in combination with the woman’s age, gives a risk percentage. So when I was 39 my triple screen results plus my age put me at a 1/200 risk of having a child with one of the three abnormalities tested. You can look at this number in one of two ways: There’s a 5% risk you child has an abnormality, or there’s a 95% chance the baby is perfectly fine. In almost any other realm these would be great odds. They are great odds. But not good enough for parents. I can understand this. A 1/200 may lead to an amnio, where the risk of miscarriage is also 5%. So a parent might opt for MaterniT21 instead. This test gives you a simple “positive” or “negative”, which is worse, actually, because now a positive convinces the mother that the sky is falling. But the test is never either positive or negative. That’s why it is a screening test, not a diagnostic test.
3. The test is only effective for people at high risk. These tests quote a specificity rate, the ability to rule out those unaffected pregnancies, of 99.9%. But mothers under 30 are already at that 99.9% rate based solely on their age. So if, like my friend, you are not of advanced maternal age (AMA) defined as older than 35, and you have no other risk factors, the test tells you nothing. “High risk” is defined by Sequenom as 1) advanced maternal age, 2) personal or family history of chromosomal abnormalities, 3) positive serum test (I’m assuming that 1/200 number or greater), 4) fetal ultrasound abnormality. (Ultrasound findings themselves have a very high false positive rate. Nuchal translucency, for example, has a 5% false positive rate.) My friend had none of these.
4. Somebody is making a lot of money, and its not me. Or you. Sequenom is currently quoting $235 – $475 out-of-pocket, and some insurance programs will cover the test in some cases. Sequenom is, however, charging $2,762 dollars for the test, which somebody is paying. Now, some people say it’s worth every penny for the peace of mind. But it doesn’t guarantee peace of mind, and the peace of mind it does bring is coming to people with insurance or who can afford the out-of-pocket.
Many reading this post will say, “but Shirie, I just know I wouldn’t be able to sleep at night unless I absolutely know for sure that everything is fine, and no amount of money or testing is too much for that assurance.” To which I say “Just understand what you are signing up for. My second child had perfect scores on every imaginable test, and still ended up with a congenital abnormality. Life doesn’t give you these kinds of assurances.”