Accuracy of Down syndrome blood test remains unknown

Non Invasive Prenatal Screening NIPS BrandsIt is coming up on two years since Non-Invasive Prenatal Screening (NIPS) was introduced. The New England Journal of Medicine recently featured an article raising concern over how the new blood test for Down syndrome is being used. It cautions against the new test being used in the general population. 

In October 2011, Sequenom launched its MaterniT21 test into select markets. Since that time, other NIPS labs–Ariosa, Verinata, and Natera–have entered the market. NIPS uses samples of the mothers blood to identify cell-free DNA to assess the likelihood of the fetus having Down syndrome. Originally promised to be diagnostic, the tests remain a screening test with false-positives and false negatives.

Professional medical societies for obstetrics, genetic counseling, and prenatal diagnosis have issued statements saying the test is only recognized for women already considered at a higher chance for having a child with Down syndrome. This is because almost all of the research by the NIPS labs has been of the high-risk population. But, some practitioners are ignoring the recommendations and offering NIPS to all patients.

In response, the New England Journal of Medicine featured an article saying the new blood test should not be offered to all expectant mothers.

The authors are with Harvard’s Safra Center for Ethics, Harvard’s Department of Health Policy and Public Health, and Massachusetts General Hospital’s Department of OB/Gyn. They state:

[T]he diffusion of [NIPS] into routine prenatal care may be occurring too quickly. Professional societies do not recommend these tests for normal-risk pregnancies because their clinical utility in the general population is not well established. … [NIPS] testing seems to be drifting into routine practice ahead of the evidence.

The article shares a data point I had failed to appreciate to support its statement that “[w]ithout additional evidence, however, the clinical utility of [NIPS] remains uncertain”

The studies the NIPS labs conducted were of populations with unnaturally high incidence rates for Down syndrome.

Down syndrome is a rare condition. For example, the baseline chance a 35-year old woman has of having a child with Down syndrome is 1-in-200, or a 0.5% chance. However, the study Sequenom used as its basis for introducing NIPS into the market had a prevalence rate as high as 1-in-8, a 12.5% chance, in the samples tested. That is a leap in likelihood going from 1-in-200 to 1-in-8.

Why I may have missed this (and you as well), is that the NIPS labs did not exactly highlight this unnaturally high prevalence rate:

It is worrisome that some laboratories that performed validation tests may have been aware that the samples included high proportions of specimens with known aneuploidies — but that this isn’t always made clear in the studies’ descriptions.

NIPS is marketed as having a greater than 99% ability to detect those pregnancies actually carrying a child with Down syndrome, and an even greater than 99% ability to rule out those pregnancies that are not carrying a child with Down syndrome. But, the meaning of those percentages change drastically depending on how common the condition is in the tested population.

At a prevalence of 1-in-8, those 99% detection rates means a positive test is 97.94% positive, and a negative is 99.99% negative. Meaning, a “positive” NIPS result is darn near positive at that high of an incidence rate of 1-in-8. Except, you only get to that high of an incidence rate in the natural population if you’re testing mothers who are over the age of 45.

For 35-year old moms, who have a 1-in-200 chance of having a child with Down syndrome, those 99% rates for NIPS plummet in how likely they are actually positive or negative. A positive NIPS test of a population with a 1-in-200 baseline chance for having a child with Down syndrome is really only 62.59% likely–not 97%.

Now consider that for moms in their twenties, their baseline chance is more than 1-in-1,000–a five times less likely chance than a 35-year old. That would make a NIPS test in that population fall even further in its ability to predict whether the pregnancy is positive for Down syndrome.

Yet, physicians are ignoring their professional society statements and the NIPS labs, as the Journal article notes, are “build[ing] consumer demand … by aggressively marketing the tests, emphasizing data that do not answer key questions.” (emphasis added).

As a result, yet another medical journal has published a statement cautioning against the use of NIPS in the general population and calling into question the value of a NIPS test result in a low-risk population.

Comments

  1. And still no answer as to the therapeutic benefit of the test.

  2. Hi. Thank you so much for these articles. I recently had the verify test done, and at the moment I am assuming that it is the same as the harmany tests and others listed. If not please let me know. My test came back positive for down’s syndrome. During my visit with the doctor for the results, my doctor urged me to get an amniocentesis to confirm, but also noted that the test results were very accurate in diagnosing down’s even if I choose not to do the amnio. She also mentioned several times that abortion was optional. Even when I told her that I could not nor ever consent to an abortion.
    Can you tell me is this test a diagnostic test. I am hoping for a false positive. But also preparing for a child with Down Syndrome. My other concern is that how many mothers out there are considering this test as an infinitive diagnostic test. Especially when advertised that it replaces invasive procedures for this type of diagnosing. How many of these women are aborting based on these results? I know you don’t have that answer. But for once can our medical community be honest about exactly what it representing? Thank you

    • Thank you for sharing your experience. Unfortunately, you are correct. A report in December 2014 in the Boston Globe reported that one lab found 6% of its screen positives were aborted without a diagnostic test, so that is happening. Since you have decided to continue your pregnancy, I hope you find Diagnosis to Delivery a helpful resource along with the other information resources at downsyndromepregnancy.org.

    • rachel jordan says

      I to hoped for a false positive. but it was confirmed through amnio that he was down syndrome. i cried and mourned my baby i felt i lost. I to could not do abortion, and so here is my son hes my life and hes so smart! its so scary at first but you will see it is not a bad thing. Down syndrome is not a rare condition. it is very common. i was only 30 when i had my son. the amnio is not diagnostic! they stick a long needle in your belly through the sac to your amniotic fluid. they take some of the fluid and test it for trisomy 21.. not painful but it has its risks.l

      • Thank you for sharing your experience. To clarify, while false positives and false negatives even happen with amnio, they are so rare that amnio is considered a diagnostic result and the gold standard for prenatal diagnosis (i.e. more accurate than CVS).

  3. Hi Mark
    Thank you for your website, the information is very helpful and your family is beautiful and inspiring!

    I have some concerns over the nips.
    I am very confused over the ability to trust these test results. I am 42 years old and pregnant with my second child. I took the panorama test and came back as low risk. I have a hard time trusting these results because the maternal fetal medicine area is always sure to finish each sentence with “this test is not 100%”. With my last pregnancy I had a low risk panorama test but then found soft markers on level 2 ultrasound. Being the type of person that “needs to know” I opted for the amino. I really don’t want to go through that again but I feel like they discount the test every chance they get ( not sure if the are just constantly adding disclaimer for legal reasons). Is a clear level 2 ultrasound and a low risk harmony something to trust or do you really have to do an amino if you “need to know”.
    Appreciate your feedback.

    • I’m essentially cutting and pasting a response at this post that I had to Dawn–who was 42 and had a screen-negative NIPS result: Dawn–Professional guidelines advise that NIPS results, like [Panorama], remain screening tests and certainty can only be provided by diagnostic testing. Further, the NSGC’s fact sheet on NIPS explains that because CVS tests the same genetic materials as NIPS, it is not as confirmatory of a diagnostic test as amniocentesis. That said, if I’ve done my Bayesian analysis correctly (which you can read about at this post), your chance of having a false screen-negative, given your age, is 0.005% or roughly 1-in-20,000 [however, I’m a lawyer, and the joke is we go to law school because we’re not good at math; I would have what is called your “negative predictive value” confirmed by a genetic counselor or medical geneticist]. All this being said, an amnio remains the only diagnostic test to give the best certainty. However, since you have a screen-negative NIPS result, I don’t know if that would be covered by your insurance, so you may be going out-of-pocket to pay for it. If that is the case, I would consider running another NIPS test. Again, something that won’t likely be covered, but that may give you the certainty you desire since the chance of two false negatives would be even rarer. Plus, as the pregnancy progresses, there is more cell free DNA from the pregnancy in your bloodstream to be tested providing more accurate results. Finally, as this post covers, though, there remain conditions that NIPS cannot test for. If you had other traditional screening (NT or quad), I would consider those results as well and determine whether you should undergo an amnio. [Check the response at the linked post for links to the other references since they won’t copy over].

  4. Hi just a few weeks ago I had a Down syndrome screening done, I am 17 wks pregnant. Yesterday I found out that the results came back positive. I am 30 yrs old, first child. My dr said that these screening tests are not accurate and will schedule a more accurate test at another hospital and she also said she has seen hundreds of false positives. I am freaking out and baffeled as to why this screening is even performed if it is not accurate. It’s plain torture.

    • Tiffany–it sounds like this wasn’t explained sufficiently before you accepted the screening test. You can ask to talk to a genetic counselor to explain screening results in greater detail and then advise you so you can decide whether you want any further testing done. I hope you get the information and support you need.

Trackbacks

  1. […] I featured an article from the New England Journal of Medicine raising concerns about Non-Invasive Prenatal […]

  2. […] professional medical organization that has issued a statement on NIPS and by physicians in the New England Journal of Medicine over a year […]

  3. […] at Harvard and Massachusetts General Hospital raised concerns about NIPS moving ahead of the evidence: “T]he diffusion of [NIPS] into routine prenatal care […]

  4. […] New England Journal of Medicine noted that the reported accuracy rates of 99% for detecting pregnancies actually carrying a child with […]