SEARCH

SEARCH BY CITATION

We are in the midst of a paradigm shift in the way that prenatal screening and diagnosis are performed around the world. This change is occurring in real time at an extremely rapid pace that is unprecedented in the history of prenatal care.

The quest for a less invasive approach to prenatal diagnosis has been the focus of much research over recent decades. Initially this was focused on the potential of fetal cells in maternal blood.[1] Intact fetal cells have been difficult to isolate and analyze, although research in this area continues.[2] When Lo and colleagues identified male cell-free fetal DNA (cffDNA) in the blood of pregnant women in 1997, this allowed the dream to be realized.[3] It then took less than 5 years for this observation to be translated into the management of pregnancies at high risk of sex-linked diseases[4] and, shortly thereafter, hemolytic disease of the newborn.[5] Now, less than 15 years later, and much to the satisfaction of women and health professionals,[6] noninvasive fetal Rhesus D disease typing is routinely offered in Europe to facilitate targeted immunoprophylaxis by using anti-D immunoglobulin and reducing maternal exposure to human blood products.[7] Identification of Y-chromosome specific sequences is also standard practice in many European countries for fetal sex determination in pregnancies at risk of serious sex-linked disorders and congenital adrenal hyperplasia, as it reduces the need for invasive diagnostic procedures,[8] which is much welcomed by all stakeholders.[9]

However, it took major advances in DNA sequencing technology that occurred in the first decade of the new millennium before it was possible to detect fetal chromosomal abnormalities without either isolating pure fetal DNA or amplifying a uniquely fetal sequence in maternal blood. Furthermore, it took a significant change in the way sequencing results were interpreted to advance the field. Rather than to simply sequence the DNA, in 2007, two separate groups hypothesized that if you sequenced, mapped, and counted DNA molecules relative to a reference standard, fetal trisomy should be detectable even in the presence of large amounts of maternal DNA.[10, 11] By 2008 proof of principle experiments had been published,[12, 13] which then gave way to large-scale blinded clinical trials of noninvasively detecting aneuploidy by performing massively parallel sequencing of maternal plasma DNA.

On October 17, 2011, testing became clinically and commercially available in the USA (it had become available earlier in mainland China). This was followed by multiple publications, several professional society recommendations, and a logarithmic uptake in the number of tests ordered. The reason why noninvasive prenatal testing (NIPT) represents a paradigm shift is that it changes the algorithm of screening followed by invasive testing that has been in practice worldwide for the last 30 years. Even in the first year of NIPT's integration into clinical care many medical centers are witnessing a significant decline in the number of invasive procedures being performed for aneuploidy.[14, 15] In addition, every aspect of the current standard of care is being questioned – for example, do we still need to measure maternal serum biomarkers, and what is the place of nuchal translucency measurement?

To provide the readers of Prenatal Diagnosis with an overview of this rapidly changing field we have taken the unprecedented step of having two back to back special topic issues on NIPT by using maternal plasma DNA. With the aid of many highly respected researchers and thought leaders in the field, in this issue and the one that follows in July, we reflect on the revolution that has occurred in prenatal testing and diagnosis. The main focus is on NIPT for the detection of aneuploidy. The developments in this area have been dramatic and already affect clinical practice. Unusually, much of the research has been largely driven by the commercial sector, because this represents a huge potential market.

In the June issue, we start with an external objective assessment performed by the California Technology Assessment Forum that reviews the technology and standards that should be required for implementation into clinical practice.[16] The guidelines of the California Technology Assessment Forum are used by many insurance plans in the USA to determine whether new tests should be covered. Walsh and Goldberg conclude that, whether using a whole genome or targeted approach, laboratory standards have largely been met for the use of NIPT to detect fetal aneuploidy in high-risk populations. This view has also been endorsed by a variety of national and international professional societies including the International Society of Prenatal Diagnosis,[17] the American College of Obstetricians and Gynecologists and the Society for Maternal Fetal Medicine,[18] the Society of Obstetricians and Gynecologists of Canada, and the American College of Medical Genetics and Genomics.[19]

In the USA, four companies are currently offering NIPT on a clinical basis. In yet another unprecedented aspect of this field, all of the companies are engaged in some sort of litigation against each other. Although the litigation does not currently stop the testing from being offered, the intellectual property rights associated with this field are complex and have certainly provided many billable hours for the lawyers involved. In this issue, Agarwal et al. explain the commercial landscape underpinning the delivery of NIPT in the USA and provide a helpful overview and background to the salient issues.[20] Whether or not the intellectual property disputes will ultimately have an impact on implementation into the public sector remains to be seen.

To date, most publications reporting on cffDNA analysis for aneuploidy have focused on the technical aspects, but as Skirton and Patch point out in the June issue, there are other non-laboratory issues to address when moving to clinical implementation.[21] In their systematic review they describe both the positive views of women and health professionals to the improved safety of NIPT, but they also highlight the different opinions that these two groups have with regard to testing. They also raise ethical and societal concerns that will require further consultation with health care policy makers. They also highlight the need to ensure provision of both time and information prior to testing to ensure maintenance of informed consent and indicate that this may require significant health professional education.

With regard to the technical aspects of NIPT, some companies offer whole genome sequencing for whole chromosome aneuploidy, whereas others offer targeted approaches that focus on the chromosomes of interest (e.g. 13, 18, and 21). In the June issue, Boon and Faas offer an objective comparison of these two approaches. They conclude by concurring with the view that high throughput may be facilitated using a targeted approach, but they question whether the lower cost of sequencing with a reduced number of tags will actually be realized in practice.[22]

A highlight of these special topic issues is the description of translation of research to clinical practice as we report the performance of NIPT for aneuploidy in the clinical, as opposed to the research, setting. Here we publish a small study of NIPT in women at low risk for aneuploidy using a targeted approach[23] and a larger one by using massively parallel whole genome sequencing in over 5000 women.[24] Both of these studies confirm performances previously reported in demonstration projects. Regardless of the approach used, in around 1 : 200 cases there is discordance between the NIPT results and the fetal metaphase karyotype obtained by the confirmatory invasive procedure. Here we publish several reports that describe potential underlying reasons for the discordance, such as confined placental mosaicism,[25, 26] maternal chromosomal abnormalities,[26, 27] and in one very unusual case, the presence of a maternal solid tumor.[28] These reports indicate that the cases are not really ‘false positives’; they are detecting unexpected findings because of the increased sensitivity of the technology.

Although the early literature focused on the detection of the major autosomal trisomies, with refinements in the bioinformatics analyses, it is clear that NIPT can also be used for the detection of sex chromosome abnormalities using either targeted[29] or whole genome approaches.[24, 30, 31] Bioinformatics manipulation can also improve detection rates for other aneuploidies, such as trisomy 13 by adjusting for factors such as guanine cytosine content.[31, 32] Other unbalanced chromosomal rearrangements, such as large duplications and deletions that broaden the potential scope of NIPT are also reported here,[33] although at present most reports describe the need for greater depth of sequencing[26] or the application of targeted sequencing.[29] The potential for testing in twin pregnancies is also being realised.[34]

In the past 18 months, over 100,000 DNA tests have been performed for aneuploidy across the globe, and factors that affect test performance are becoming clearer. The relative proportion of cffDNA, or the fetal fraction, is critical.[35, 36] This is directly related to both maternal weight and gestational age.[36] An alternative approach to measuring fetal fraction can be delivered through quantification of RASSF1A. This imprinted gene is a uniquely placental marker that is elevated in pre-eclampsia.[37]

Although most prenatal testing is currently directed at the detection of aneuploidy, a small but significant number of families seek noninvasive antenatal diagnosis because of a risk of a monogenic disorder. However, development in this area has been slow, largely because of the lack of commercial drive and the need for tests to be developed on an individual patient or disease-specific basis, which does not lend itself to high throughput delivery. Lench and colleagues explain how many of these challenges can be overcome with the increasing availability of next generation sequencing in public health sector laboratories.[38]

Looking towards the future, Snyder et al. clearly explain how genome-wide inherited and de novo variations in the fetal DNA sequence can be determined without incurring risks to the mother or fetus.[39] If noninvasive determination of the entire fetal genomic sequence becomes clinically available, there will be a significant increase in the number of ethical issues that arise. These authors reflect on the many technical and translational challenges that need to be overcome before introduction into clinical practice. This will hopefully give us enough time to address some of the ethical and moral dilemmas such an approach to prenatal diagnosis may pose.

Today the majority of prenatal tests for monogenic and chromosomal disorders are performed to facilitate parental choice to continue or terminate an affected pregnancy. Fetal treatment options remain very limited. However, advances in gene therapy make in utero treatment increasingly possible,[40] and early diagnosis based on the analysis of cffDNA will enable optimal timing for any such treatment. This may not solely be limited to monogenic disorders, as expounded by Guedj and Bianchi, who review the information on fetal brain pathology in Down syndrome and argue that there is a theoretical opportunity to improve neurogenesis and brain morphogenesis by in utero treatment.[41]

As demonstrated by the articles in these two issues, NIPT is a very rapidly developing and exciting area. Here we have aimed to capture the main advances in the field and highlighted areas of future development. There is little doubt that tests based on cffDNA provide a readily accessible and generally safer option for prenatal testing that can be offered from 10 until 40 weeks of gestation. The challenge now is to translate this technology into practice that is accessible to all pregnant women, and in an ethical way that preserves informed parental choice, while not increasing overall costs to the heath care system.

REFERENCES

  1. Top of page
  2. REFERENCES