Non-Invasive Prenatal Genetic Diagnosis – An Update

Today researchers employed by Sequenom published the results of a trial of their method of detecting fetal trisomy 21 (Down Syndrome) from the blood of pregnant women in the American Journal of Obstetrics. I have not yet been able to read the article, having been unwilling, thus far, to spend $30 to download it, but secondary sources report that the trial was a success.

The company looked at blood plasma from 480 women at higher than normal risk for having fetuses with trisomy 21. 31 of the samples were removed from the analysis for various technical reasons. Of the remaining 449 samples, the Sequenom test correctly identified all 39 cases of trisomy 21 (a sensitivity of 100%) and incorrectly identified only one of the 410 non-cases as positive (a specificity of 99.7%).

As discussed before in this blog, the method sequences the cell free DNA in the maternal serum and, basically, counts how many bits of chromosome 21 are present. This report, along with the January BMJ publication of the large trial by Dennis Lo’s group in Hong Kong, seems to me to make it all the more likely that non-invasive prenatal diagnosis of aneuploidies will be clinically available in the US soon . . . by 2012? But

will it be regulated by the FDA?
will it be paid for by insurers and Medicaid?
will it expand to Mendelian diseases and traits?
will (and should) we try to regulate the diseases and traits for which it is used?

We need to be exploring these questions!

6 Responses to Non-Invasive Prenatal Genetic Diagnosis – An Update
  1. The article is available online for free, below is the link and thanks for your insight.


  2. The problem with a non-invasive diagnostic test for Down syndrome is the assumption that women aren’t concerned about non-Down syndrome chromosome abnormalities. At present women who are at increased risk for Down syndrome on non-invasive screening tests, and who choose to then have invasive diagnostic testing, are actually having a test that diagnoses all chromosome abnormalities, not just Down syndrome. In fact, nearly half the abnormalities that are diagnosed in this group of women are something other than Down syndrome, including not only other aneuploidies, but also structural abnormalities including deletions and unbalanced rearrangements. If this same group of women were to have NIPD for Down syndrome instead of a karyotype (which is what Dennis Lo suggests in this latest publication), the only chromosome abnormality that will be diagnosed is Down syndrome. This means that potentially there will be other chromosome abnormalities present that will go undetected. Of course with NIPD there will likely be a larger number of women at risk of Down syndrome taking up diagnostic testing, and this may be seen as a good thing, but if women don’t realise that there is a trade off involved, that other chromosome abnormalities may no longer be detected, then it is hardly informed decision making. (Susman et al (2010), Genetics in Medicine:May 2010,12(5) pp 298-303)

    It is also slightly ironic that at the same time as NIPD for Down syndrome is being promoted as the best thing since sliced bread, that the use of arrays for prenatal diagnosis is being offered to women who have abnormal findings on ultrasound.

    It all begs the question – why do we offer prenatal diagnosis? Is it to detect as many Down syndrome pregnancies as possible or is it to detect as many severe abnormalities as possible?


  3. Thanks for your comment.

    I agree that NIPD would not be an important change if it only looked for Downs syndrome and could even lead to less genetic testing in such a case. But I don’t think it will stop there. The current efforts are starting with Downs (and other aneuploidies – the Quake group has consistently looked for trisomy 13 and 18 as well), but in the not very long run, they should be able to get lots of single gene information. It won’t be exactly the same as the information provided by arrays, but should be able to provide an awful lot of information . . . and without using amnio or CVS. That’s the revolution, not just a non-invasive diagnosis for Down syndrome.

    Right now under 2% of US pregnancies involve amnio or CVS; about 2/3 of pregnancies in California involve Downs screening. You wouldn’t have to add much to the trisomies for NIPD to provide information to a lot more women than existing levels of amnio and CVS can provide.

    I am curious, though. If a women gets a positive screening test for Down’s (or another trisomy) and proceeds to amnio because of it, what kind of informed consent does she get for all the non-trisomies that might be found through her sample? Do women in that situation often want to find out “everything possible” from the amnio sample even though they are getting it only because of trisomy risks?

    Hank Greely


  4. Thanks for your reply.

    In answer to your question, some women do want to find out everything possible. There is a recent publication indicating when women have a choice between karyotyping and rapid aneuploidy detection (RAD) with invasive testing, that they choose karyotyping because they want to know everything, but choose RAD because it has a shorter turnaround time.

    I also think that it is really important that the issues regarding the potential introduction of NIPD for Down syndrome alone, are discussed. Although at some future date, either the Quake or Lo groups may develop the bioinformatics neccessary to detect trisomy for chromosomes 18 and 13, it looks like Sequenom are focussed on NIPD for Down syndrome. To make an informed choice between NIPD for Down syndrome or invasive testing with a karyotype, women will need to understand what abnormalities are detected by a karyotype, and that these other chromosome abnormalities make up around 50% of the abnormal results detected from CVS and amniocentesis.

    I think we need to be discussing what might happen in the immediate future. If we can have appropriate education and informed consent procedures sorted out for an NIPD test for Down syndrome, it will set the scene for the more complex questions that will arise when other conditions are added to the panel.


  5. Marleen – you’ve convinced me. (Or, more accurately, your Genetics in Medicine article convinced me.) The ironic possibility that early stage, trisomy 21 only NIPD could lead to women getting less genetic information about their fetuses needs to be taken seriously. I do think that’s likely to be, on balance, a short term issue, but a lot of women will get pregnant in that short term. This is one more important issue to add to the many issues around NIPD that we need to start discussing. (And, in the long run, NIPD probably won’t be able to do everything that karyotyping does, though, of course, it will be able to do a lot that karyotyping doesn’t do.)

    I hadn’t run across your work on NIPD before – glad to see that there is work being done on it in Australia.


  6. Thanks for reading my paper and I’m glad it has convinced you. The question that this raises is how are women going to be properly informed about all these chromosome conditions and how will they make decisions about what they want prenatal testing for. With our current system they really only find out about Down syndrome. This question will be even more pertinent when additional conditions get added to NIPD panels.

    The conclusion from the GIM paper has led to my PhD project – a questionnaire (based on the Theory of Planned Behaviour) to find out how women will make decisions about NIPD for non-severe conditions. I have completed a preliminary study to elicit beliefs from our population about testing for different types of chromosome abnormalities, and not surprisingly there was consensus about testing for severe disorders. For the main questionnaire we are therefore concentrating on four less severe conditions where there was some discordance in responses. To prevent inbuilt prejudice and misunderstanding about particular conditions, the conditions are not named- a short description is all that is provided. I am currently still recruiting women for this questionnaire and hope to have 500 responses before I start the analysis.

    The results should give us a better understanding of the range of women’s opinions about testing for less severe abnormalities and how they make those decisions. I hope that the results will stimulate discussion on NIPD that includes women’s perspectives.


Leave a Comment