NEW YORK (GenomeWeb) – In a commentary in Nature this week, Diana Bianchi, a medical geneticist at the Tufts University School of Medicine, has made a plea to the non-invasive prenatal testing field to standardize its consent and patient counseling procedures to make sure women understand that there is the potential they might learn something not only about their fetus, but also about their own health.
The issue of incidental findings in NIPT has grown as this type of testing has grown, from a novel technology just a few years ago to an increasingly used technique to screen women of advanced age or otherwise elevated risk for fetal aneuploidies. It also has the potential to replace serologic tests as a first-line screening tool.
Earlier this year, Sequenom, one of the major NIPT providers, disclosed data on a group of cases in which its test uncovered abnormal aneuploidy findings that were confirmed not to be fetal and were later linked to cases of maternal cancer.
One of these cases, Eunice Lee, shared the story of her own cancer diagnosis prompted by these abnormal NIPT results at the Scripps Future of Genomic Medicine Conference in March.
Her story raised practical and ethical questions about how women are consented and counseled as part of receiving these tests, which analyze the DNA of both the fetus and the mother — something that may not be readily understood.
Based on these issues, Bianchi has reviewed consent forms from the major testing companies in the field and has found wide variation and some troubling details.
As the field is poised to grow and diversify, she has called on test providers to rethink their consent forms to prevent unwarranted confusion or anxiety, and on professional societies, such as the American College of Medical Genetics and Genomics, the American College of Obstetricians and Gynecologists, and the Society for Maternal-Fetal Medicine to take the lead on providing education and clinical guidance.
Bianchi spoke with GenomeWeb this week to discuss her findings and recommendations.
This is something we've been more aware of since Sequenom presented on their cases of NIPT results linked to maternal cancer, but for you, at what point did it start to click that there may be problematic variation in the consent procedures?
Since the beginning, when NIPT became clinically available, I've been very interested in the biological reasons for the so-called "false positive" cases. If you recall, the concern was that there was something wrong with the technology, and there was an alarm raised. In fact, one of the papers actually had the word 'alarm' in the title. So, along with my colleagues, we've been tracking the reasons for these test results that are discordant with the fetal karyotype, and because we've been collecting preliminary data for a grant application, I've become aware of many such cases. That's what prompted me to start looking at the consent forms.
The other thing that happened was that in the beginning it was the big six -- the four companies in theUSand two inChina-- but as of late 2014, this testing became decentralized, so now there are more companies involved, academic laboratories are licensing the technology, etc. Thus, the consent forms have changed over time, and I've been reading the different consent forms as new companies have gotten involved in the noninvasive prenatal testing space.
There has been a lot of attention paid to the Sequenom maternal cancer cases, but it's not just cancer that's a potential concern. Can you run through the different things that a woman could learn from the results of these tests regarding her own health?
The potential for maternal cancer diagnosis is very attention grabbing. These cases are the most dramatic because they are potentially actionable for the mother. But I think for me, a big issue — and hopefully this will be clarified by the Nature commentary — is the fact that unlike in other areas of genetics, [in NIPT] you are testing two people, not one. When I had my whole genome sequenced, it was just me that I needed to worry about. In pregnancy, however, most women don't realize that they are going to get information on their fetus, but potentially from themselves as well. So, there's a small but real possibility that something will be discovered in the maternal DNA fragments that are sequenced.
To answer your question, these technologies are looking at whole chromosomes or parts of chromosomes, so you could, for example, and we know of these cases, find out that a mother has mosaicism for an autosomal aneuploidy.
In terms of maternal incidental findings, sex chromosomes are probably the most frequent. As a medical geneticist, one of the most interesting things about the maternal findings is discovering how prevalent sex chromosome aneuploidies are in fertile women. In the past, when we counseled girls or women who were diagnosed with a sex chromosome abnormality, we usually counseled them that this could impact their fertility. But now [we are] finding evidence of so many more women who are fertile with that condition. So, noninvasive prenatal testing is unexpectedly adding information to our knowledge of the natural history in supposedly well-known genetic conditions such as the sex chromosome abnormalities.
Also, there are the microdeletions. We don't have a lot of information yet on incidental detection of maternal DiGeorge syndrome. However, in oral presentations at meetings, I've heard speakers present [unpublished] data on the unexpected detection of DiGeorge syndrome in a parent.
You mention in your commentary that current consent forms "rarely" mention the possibility of findings concerning the mother's health. Do we know exactly how each company doing this type of testing is consenting subjects?
You can manually review the consent forms, which is what we tried to do. We didn't get all of them, because some companies will only let you have the consent form if you are a client who is going to order a test. The whole point of the commentary, however, was a plea, essentially, for the laboratories that are providing this testing to re-examine their consent forms in light of evolving information on maternal incidental findings.
I don't fault the screening laboratories, as some of the relatively rare maternal findings, like cancer, weren't observed until they started doing hundreds or thousands of tests, because cancer is so rare in women of reproductive age. What was most concerning to me was that in several situations, the woman herself didn't even sign the consent forms.
The doctor signs it?
Yes, the doctor signs it and just says 'the woman has been counseled.' That was the first point I tried to make in the commentary. Women should receive pretest counseling and they should sign the consent form indicating that they have been counseled.
So what's the potential harm if patients don't fully understand that they might get an abnormal result that implicates their own health, and then if they do receive such a result, but haven't been properly advised of what it means?
Well, I think the biggest harm is that they potentially don't understand that the test is sequencing the maternal DNA as well as the placental DNA. They might misunderstand the implications of a positive test result, especially if they don't have a follow-up confirmatory diagnostic test on the fetus. In that situation, they might incorrectly think the fetus has a chromosome anomaly and take irrevocable action related to that.
And the decision whether or not to receive results about their own health, is that something you think is important as well, that they should be able to opt in or opt out of that?
I do believe women should be given a choice. Some are going to say they don't want to know, but they need to be given a choice, and part of making that choice is making sure women understand that the tests are looking not only at the fetus, but also analyze the maternal DNA.
So what does the optimal consent form for NIPT look like?
Part of the issue is that as this testing disseminates, it's not going to be genetic counselors doing the counseling. I know there are people inBostonandNew Yorkwho are getting the test without really getting any counseling. Their doctors are ordering it, and if they're over 35 they're just checking it off.
So, it's a big concern. It's really important that there is time for the pre-test counseling. It's difficult, but that's why I think we should be thinking a little more proactively about not necessarily offering counseling in the old "one-on-one" model, but at least having well-curated information endorsed by professional societies to be used in a variety of settings, and in a variety of languages, and making it accessible even to people who don't have a computer or a smartphone.
This education obviously needs to include all of the providers as well — so not just the physicians but the nursing staff, physicians assistants, whoever is really upfront with the patient. And then, the consent form needs to be provided in a variety of languages. Again, I think this is a great opportunity for the patient herself to sign the consent form rather than the physician.
This is not a simple straightforward test. It is more complex than some providers realize. I [also] think the current standard of care, serum biochemical testing along with the nuchal translucency ultrasound measurement is also complex. Pretest counseling is not always being done for that either. But, this is an opportunity to start fresh and use noninvasive prenatal testing as an example for all of genetic testing.
In a way, NIPT is the "poster child" for the future of genomic medicine. Just look at the numbers of tests performed in the last three years. I used GenomeWeb as my data source. There are 196 articles on NIPT, and I went through every one to find the number of reported tests per quarter. At the time [prior to their IPO] Natera didn't report their numbers, so I didn't include them, but I was able to get enough information that seemed consistent through the reports. If you add up the numbers, it was 1,503,000 tests worldwide that had been done between 2011 up until December 2014. We won't be able to get the number in 2015 because of the decentralization of laboratory providers.
Overall, the realization that these tests could also be a source of maternal health information, specifically cancer detection, holds this huge positive potential, and we know now that many of these companies have plans to adapt their technologies specifically for early cancer detection. To do that they'll have to do more research and publish validity data showing the tests can do this accurately. But in the meantime, more and more women are being tested for fetal chromosomal abnormalities, and by virtue of how these technologies work, more cancers and other disorders will be detected, right?
You and I were both at the session [at the Future of Genomic Medicine conference] where the Sequenom clinical laboratory director was discussing how difficult it was to report the [cancer] information. I think it's really important to communicate that this test is not designed nor validated for the detection of cancer. In this situation cancer is truly an incidental finding, and it's a dramatic incidental finding, but it's still extremely rare.
That's why the companies are really grappling with how to report this information, because more likely than not, they are going to have false positives for cancer. If you have a pregnant woman with unusual NIPT results, to whom are you going to make the phone call to say we do think your patient may have cancer? It's not clear yet exactly what the cancer signature might be. It is truly an area for research to try to determine exactly what the pattern of chaotic DNA findings is that is likely to be associated with a tumor.
Any other ways you are hoping to work on studying or addressing this issue? You mention that The Netherlands has a large trial to comprehensively study NIPT. How can we get something like that going in the US?
There was a meeting on the possibility of developing an NIPT registry for the US at the American College of Obstetrics and Gynecology meeting earlier this month in San Francisco. Meanwhile, there is a significant need for objective, ideally third party-validated information on what the underlying biological reasons are for discordant NIPT results. The professional societies are very clear in recommending a diagnostic procedure, whether it's amniocentesis, CVS, or a neonatal karyotype, [and] that's really what needs to be done when there is an abnormal NIPT result.
We are hoping that in most cases that is going to get done. The fetal karyotype will be either concordant or discordant with the NIPT results. And in the setting where the results are discordant, we need objective information to be able to counsel women as to what's going on. For example, we published a paper looking at the sex chromosome abnormalities, because the test performance in that area is not as good as with the autosomal aneuploidies. The point is, there appear to be biological reasons for the different performance for sex chromosome testing. If you get an NIPT result of 45,X and the fetus is confirmed to be 46,XX, we showed that there was a statistically significant association with advanced maternal age. The biological explanation for the discordant positive result of 45,X may be due to the known phenomenon of somatic loss of a single X chromosome that occurs normally as a women ages.
It can be very reassuring to women to know that there is an explanation for the result and that their baby is well … but all these complexities don't lend themselves to a quick in-the-office consultation.
All the more reason to make sure women know about the possibilities beforehand?
Yes, at least about the possibilities, and then they've been forewarned if they are in the rare situation of getting one of these discordant results.
What I would envision in the future ideally would be a combination of pretest counseling and then something that is web-based for women after testing, so if you have an NIPT result A, and your fetus is normal based on follow-up test results, then the possible reasons for this are X,Y, and Z, and here's what you should talk to your doctor about in terms of further management.
It's only an idea right now, but that's why we've started with the grant application, because we really have to have "third party" validated information. It is an idea that hopefully the commercial laboratories companies will support, but in the ideal world, it should really be government funded.