Abstract
After karyotyping invasively obtained fetal material for decades, the field of prenatal genetic care has changed tremendously since the turn of the century. The introduction of novel technologies and strategies went along with concerns and debates, in which key issues were costs, the finding of variants of unknown or uncertain clinical relevance, commercialization and ethical and social issues. At present, there is an explosion of new genomic technologies, which need critical assessment prior to implementation, especially in the prenatal field. The key issues of the debates we had in the past will again play a major role in guiding us toward careful implementation of these new techniques in future.
1. Introduction
In the past 15 years, the landscape of prenatal genetic analysis has changed tremendously. After karyotyping fetal material, invasively obtained via amniocentesis (AC) or chorionic villi sampling (CVS) (below referred to as ‘routine karyotyping’), for decades, shortly after the turn of the century, rapid aneuploidy detection (RAD), to detect the most common aneuploidies of the chromosomes 13, 18, 21, X and Y was introduced. RAD turned out to be very reliable, relatively cheap and, most importantly, fast: results can be obtained within 24 – 48 h, compared to 10 – 21 days for routine karyotyping. Already shortly after its introduction there were many international discussions whether or not RAD could be offered as stand-alone test for pregnancies with a relatively low risk. Even though a consensus in this discussion was never reached, it was unique in a way that it was for the first time since the beginning of prenatal genetic analysis that the use of routine karyotyping for all indications was under debate. A few years later, the introduction of array analysis was a major step forward. Although not as fast as RAD (turnaround time 7 – 14 days, depending on the laboratory logistics), it has the advantage that it allows for high-resolution analysis and thus the identification of copy number losses and gains, that remained undetected with routine karyotyping (and RAD). But by far the most exciting development in the prenatal field has been the discovery of cell-free fetal DNA (cffDNA) in the maternal plasma Citation[1], allowing to study the fetal genetic status without any risk for the pregnancy. Despite the fact that the analysis of cffDNA for the detection of fetal aneuploidies is regarded a screening test and not a diagnostic test, the introduction of noninvasive prenatal testing (NIPT) has led to a major shift from diagnostic tests performed on invasively obtained fetal DNA to highly accurate screening tests performed on noninvasively obtained fetal DNA. The initial turnaround time of NIPT was up to 14 days, but by now, most providers are able to issue a result within 10 days. Interestingly, with the implementation of NIPT also another feature, new to the prenatal field, came along: commercialization.
Most new technologies are first validated and implemented in a postnatal setting and thus, when implementing new techniques in prenatal care, the technology itself is mostly not under debate: it is the application in a prenatal setting that needs specific attention. Of course, cost and cost-effectiveness is and will always remain an important point of attention with the implementation of new strategies, but it is not specific to the prenatal field, may differ from country to country and reimbursement policies may change over time. This issue is therefore not discussed here.
Apart from costs, what were the particular points of interest with the implementation of the two major developments of the past years, array analysis and NIPT, and will the discussions we had guide us in future developments?
2. Recent advances and discussions
2.1 Array analysis
Because of interpretation challenges, before offering array analysis in a prenatal setting broad experience with the interpretation of array data and knowledge of array data from a postnatal point-of-view is of utmost importance. In the postnatal setting, high-resolution array analysis is worldwide implemented, mostly as a first tier test. There is much experience and large databases, either in-house or publicly accessible, have been built, to enable classification and better interpretation of copy number variants (CNVs).
By now, the added value of array analysis in terms of unraveling the genetic cause for ultrasound anomalies is no longer under debate. A number of studies have also shown the added value in terms of the detection of clinically significant (submicroscopic) aberrations in the population of pregnant women with relatively low risk Citation[2,3]. The exact yield, however, is not easy to determine, as not all studies use the same classification criteria.
One of the major concerns when implementing array analysis in the prenatal setting was how to handle findings of unknown or uncertain clinical relevance (VOUS – variant of unknown or uncertain significance) (even though this can be largely circumvented by using targeted techniques). Even now, a few years after the introduction, this still remains an issue of concern, as these findings are obviously more challenging to deal with in the prenatal setting than in the postnatal setting. Also for this category of CNVs, it is not easy to determine the actual proportion of fetuses carrying such a CNV, again due to the fact that not all studies use the same classification criteria.
In the past, several groups have published proposals for classification, such as De Leeuw et al. in 2012 Citation[4]. Still, not in all studies data are presented uniformly. Therefore, it is difficult to compare, summarize, oversee and draw conclusions from data from different cohorts. Reasons for the nonuniformity can be:
Different minimal sizes of CNVs.
Categorization of CNVs with known inheritance patterns in the same groups as CNVs with unknown inheritance patterns. Without making a distinction, the VOUS category can include:
De novo CNVs, of which the clinical relevance remains unknown;
Inherited CNVs, of which the clinical relevance remains unknown;
CNVs of which the inheritance pattern is unknown.
3) Reporting of CNVs in genes, associated with recessive disorders. The American College of Medical Geneticists Citation[5] already recommended that reporting of recessive mutations is outside the scope of array testing. Only in the (exceptional) case the prenatal sonographic findings can fit the phenotype of the recessive trait, one can consider to report the CNV and offer further analysis to find the second hit.
4) Reporting CNVs in or containing susceptibility loci for neurocognitive disorders. Even in the postnatal setting, reporting such CNVs is under discussion. For some of them the clinical evidence is much stronger than for others, but still these remain susceptibility loci. The question is how to deal with the finding of such CNVs in the prenatal setting, when there is no phenotype to support the finding. Should these be reported and if so, how should they be classified?
5) The categorization of different types of CNVs as pathogenic.Pathogenic CNVs can be CNVs that explain the sonographic findings, and thus are causal, or not explain sonographic findings, thus are not causal, but still pathogenic.
6) Progressive insight. New data, both postnatal and prenatal, are accumulating in literature and, as already shown by Wapner et al. Citation[2], this will lead to reconsideration and in some instances reclassification of CNVs originally classified as VOUS.
Discussions on categorization and on whether or not to report a CNV will remain, and it will be very hard to reach worldwide consensus on the labels of categories. But a general adopted classification strategy is indispensable for comparing and drawing conclusions from data from different cohorts. Recently, Srebniak et al. Citation[6] proposed an extended classification strategy, taking into account the issues mentioned above. One should, however, keep in mind that the concept of VOUS is not new and not exclusively related to data obtained from array analysis.
2.2 Noninvasive prenatal testing
Even though pregnancy loss rates after an invasive procedure are low, they are not zero and depend on the experience of the operator. Therefore, it is obvious that the introduction of noninvasive prenatal diagnosis (NIPD) for monogenic traits and NIPT for the detection of the most common fetal trisomies is generally regarded as a dream coming true, as it makes prenatal genetic analysis accessible for all pregnant women without any risk for the pregnancy. New to the prenatal field was the influence of commercial providers on the introduction of NIPT. Up till recently, new technologies and strategies had always been developed and introduced within an academic setting, whereas the primary introduction of NIPT was mainly commercially driven. This lead to concerns that, as a consequence of the commercialization, the introduction of NIPT would be less careful and could be insufficiently monitored. Nevertheless, since 2011, NIPT is offered and since then has rapidly transformed the landscape of prenatal care. As with the introduction of any new technology/strategy, also for NIPT some issues were extensively under discussion. Without pretending to be complete, the most important ones were:
Ethical and social
With NIPT, fetal genetic information can be obtained with relative ease, early and without risk for the pregnancy. Moreover, its performance outweighs that of current serum screening tests Citation[7-10], resulting in a lower invasive follow-up testing rate. When offered to all pregnant women, both professionals and pregnant women might consider NIPT to be a routine procedure, possibly leading to routinization of the pretest counseling. This might lead to insufficiently informed women, for whom a result suggestive for a severely handicapped child, might be unexpected. In the light of this, one should not forget the aim of prenatal diagnosis, which is not to prevent the birth of disabled children, but to allow for autonomous reproductive choices.
2) Confined placental mosaicisms
As the cffDNA is derived from the placenta, the result of a noninvasive DNA test may not always reflect the true fetal genetic status. For NIPD, the results are not prone to confined placental mosaicisms (CPM) and these tests are considered diagnostic. On the contrary, fetal trisomy detection using DNA from the placenta is influenced by CPM, and therefore such tests are considered screening tests: the true fetal genetic status needs to be established on fetal DNA, to be obtained via invasive procedures. As some aberrations are more prone to CPM than others, whether or not a positive NIPT can be confirmed with CVS or should be confirmed with AC depends on the aberration. If CVS is chosen as conformational test, at least the mesenchymal core cells need to be studied, as these placental cells are more representative for the fetus itself than the cytotrophoblast cells. Initially, CPM was an issue of concern for the introduction of NIPT, but it is now regarded a fact that we have to and are able to deal with, more than a disadvantage that prevents the further development and implementation. Furthermore, although not yet clear, CPM might explain some fetal and perhaps even neonatal features (e.g., intrauterine growth retardation and failure to thrive) that, up till now, often remain unexplained.
3) Fetal fraction
The cffDNA is the minority of the cell-free DNA in the maternal plasma. It is obvious that there should be sufficient cffDNA in the sample to produce a result representative for the fetal genetic status. For the time being, there is no consensus on the method for estimating the fetal fraction (FF). Some NIPT providers do make estimations and have defined minimal thresholds for FF, whereas others do not. In a recent paper, Takoudes et al. Citation[11] requested NIPT to the major providers in the US, using two blood samples from non-pregnant women. Surprisingly, of the providers that estimate the FF, one reported a normal female fetus for both samples, with sufficient cffDNA in the sample. Takoudes et al. mention ‘this example raises concerns about the need for quality standards in NIPT.’ The results of the study, however, do not support the discussion that FF should always be estimated before issuing a result, but at the very most shows that estimating the FF only makes sense when an accurate method is used. In the near future, the results of large-scale studies will probably answer the question of whether estimating the FF is of added value, as in case this is true, providers that do not routinely estimate the FF will perform poorer than those who do in terms of false-negative results. Moreover, it is expected that accurate, more uniform and easier methods will become available.
4) Quality control
As yet, there are no laboratory nor counseling guidelines for NIPT. Also no external quality assessment scheme with adequate samples is available. Quality control issues often lag behind when new techniques or strategies are introduced, and it is to be expected that these gaps will be filled in the next few years.
5) The availability of NIPT to all pregnant women
The initial clinical validation studies were performed in the group of so-called ‘high-risk’ pregnant women. But results from the general population of pregnant women are now becoming available Citation[7-10], showing that the performance of NIPT as screening test outweighs that of any other screening test. Offering NIPT to all pregnant women not only causes a major shift in the uptake of NIPT, it also leads to ethical and social discussions.
6) Expanding scope
After the introduction of NIPD for specific monogenic traits, NIPT was originally validated and offered for the detection of fetal trisomies 13, 18, 21 and the sex chromosomes, the latter sometimes being optional. But by now, the scope has already expanded to information on trisomies of other chromosomes and also to microdeletion syndromes (e.g., DiGeorge/VCF syndrome) Citation[12]. This raises questions such as to which conditions should the scope be broadened, are the tests for these conditions sufficiently clinically validated, to whom should these tests be offered and how should these women be counseled? Moreover, one should realize that the low predictive value of such expanded screening options (due to the low prevalence of the traits under study) will result in more (unnecessary) invasive procedures.
As mentioned above, the technical issues will probably be solved in the next few years, but the ethical and social debates will remain and become even more important.
3. The future of prenatal genetic care
As a consequence of the rapid technological advances in the genetic field in general, in the past few years, the landscape of prenatal care has changed tremendously. This change of landscape went along with some very important debates. Apart from technical issues (that I expect to be solved in the near future), in my opinion, key words in these debates were costs, the finding of VOUS and the associated counseling challenges, commercialization and ethical and social issues. With these issues as key players, the debates not only contributed to the careful implementation of array analysis and NIPT, but also prepare the discussion and pave the way for the introduction of even more advanced technologies. Although there is an explosion of new genomic technologies, I believe that the availability of a novel technology itself is not a sufficient argument for implementation, it should be its clinical utility. Each new technology needs critical assessment prior to implementation, especially in the prenatal field, and the same key words will again play a major role.
A few years after the introduction of array analysis in the prenatal setting, VOUS and the counseling of it still remain a subject of concern. Nevertheless, exome sequencing, a technique in which VOUS, unsolicited findings and the associated counseling are of much greater concern, is implemented in postnatal care and waiting to be given the green light in prenatal care too. In a postnatal setting, there mostly is a well-defined phenotype, based on which specific gene panels can first be sequenced, before opening up the exome (which is much more prone to unsolicited findings than panel screening, the latter regarded to be ‘targeted analysis’). But in the prenatal setting, defining the exact phenotype is far more challenging, as not all features of a syndrome are prenatally recognizable. Sequencing the whole exome instead of selected gene packages may therefore be the only possibility, inevitably leading to VOUS and unsolicited findings that need to be dealt with by the counselor (not to forget challenges with regard to data management). If future parents want to use the genetic information for making a decision on continuing a pregnancy or not, turnaround times also become challenging. But in a scenario in which we consider the fetal and neonatal life as a continuum, in which birth is ‘only’ an event, prenatal testing is not merely used for making a decision on whether or not to continue a pregnancy, but also to improve pregnancy, fetal and neonatal management and pediatric outcome. This makes the implementation of exome sequencing prenatally less dependent on very short turnaround times. With the present velocity of accumulating data, experience and technological advances, I believe exome sequencing will be implemented in prenatal care within the next few years. VOUS and unsolicited findings will require specifically trained counselors, but the debates we had and knowledge we gained with regard to the VOUS and unsolicited findings in array analysis will certainly contribute to its careful implementation. Even in the NIPT era, whether or not to offer a prenatal genetic test that needs to be carried out on invasively obtained material in my opinion is not a valid question, as, according to recent data Citation[13], the loss rate that is associated with an invasive procedure is only 0.01 – 0.02%.
With NIPT, the concept of commercialization came along in the prenatal genetic field, a concept that should not merely be regarded as negative. I think it is fair to acknowledge that without the influence of commercial providers, NIPT would not yet have been introduced as broadly as it is now. In the near future, I expect a continuing important role of commercial partners and a close relationship between commercial providers and prenatal societies will be essential to ensure that new version of tests are not only introduced, but carefully implemented, with specific attention for quality, counseling, ethical and social issues. Providers should make sure their test is offered to pregnant women via well-educated physicians and that women are correctly and sufficiently informed before autonomously choosing to have a specific test or not. Prenatal societies can and should play an important role in this.
NIPT is now evolving to fetal molecular karyotyping and proof-of-principle studies already showed the feasibility of fetal whole genome sequencing, and not to forget the determination of the fetal methylomic Citation[14] and transcriptomic profiles Citation[15], using cffDNA and cffRNA. In addition, there is increasing evidence that cffDNA can be used as biomarker, to provide information about the placenta and to be used to predict clinical problems (as reviewed in Citation[16]). It is very likely that genome-wide fetal analysis using cffDNA will play an important role in prenatal care in the future, perhaps replacing exome sequencing on invasively obtained fetal DNA. The ethical, social and legal discussions that go along with this follow the line of those already in place and I completely agree with Dennis Lo in his excellent review Citation[17]: ‘It is thus timely and important that discussion on the ethical, social and legal implications of NIPT be carried out and that these catch up with the technological progress.’
Declaration of interest
The author has no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.
Bibliography
- Lo YM, Corbetta N, Chamberlain PF, et al. Presence of fetal DNA in maternal plasma and serum. Lancet 1997;350(9076):485-7
- Wapner RJ, Martin CL, Levy B, et al. Chromosomal microarray versus karyotyping for prenatal diagnosis. N Engl J Med 2012;367(23):2175-84
- Srebniak MI, Van Opstal D, Joosten M, et al. Whole genome array as a first-line cytogenetic test in prenatal diagnosis. Ultrasound Obstet Gynecol 2014;45(4):363-72
- de Leeuw N, Dijkhuizen T, Hehir-Kwa JY, et al. Diagnostic interpretation of array data using public databases and internet sources. Hum Mutat 2012;33(6):930-40
- Kearney HM, Thorland EC, Brown KK, et al. Working Group of the American College of Medical Genetics Laboratory Quality Assurance C: American College of Medical Genetics standards and guidelines for interpretation and reporting of postnatal constitutional copy number variants. Genet Med 2011;13:680-5
- Srebniak MI, Diderich KE, Govaerts LC, et al. Types of array findings detectable in cytogenetic diagnosis: a proposal for a generic classification. Eur J Hum Genet 2014;22(7):856-8
- Bianchi DW, Parker RL, Wentworth J, et al. DNA sequencing versus standard prenatal aneuploidy screening. N Engl J Med 2014;370(9):799-808
- Dar P, Curnow KJ, Gross SJ, et al. Clinical experience and follow-up with large scale single-nucleotide polymorphism-based noninvasive prenatal aneuploidy testing. Am J Obstet Gynecol 2014;211(5):527.e1-527.e17
- Zhang H, Gao Y, Jiang F, et al. Noninvasive prenatal testing for trisomy 21, 18 and 13 - clinical experience from 146,958 pregnancies. Ultrasound Obstet Gynecol 2015;45(5):530-8
- Norton ME, Jacobsson B, Swamy GK, et al. Cell-free DNA analysis for noninvasive examination of trisomy. N Engl J Med 2015;372:1589-97
- Takoudes T and Hamar B. Performance of non-invasive prenatal testing when fetal cell-free DNA is absent. Ultrasound Obstet Gynecol 2015;45(1):112
- Wapner RJ, Babiarz JE, Levy B, et al. Expanding the scope of noninvasive prenatal testing: detection of fetal microdeletion syndromes. Am J Obstet Gynecol 2014;212(3):332.e1-9
- Akolekar R, Beta J, Picciarelli G, et al. Procedure-related risk of miscarriage following amniocentesis and chorionic villus sampling: a systematic review and meta-analysis. Ultrasound Obstet Gynecol 2015;45(1):16-26
- Lun FM, Chiu RW, Sun K, et al. Noninvasive prenatal methylomic analysis by genomewide bisulfite sequencing of maternal plasma DNA. Clin Chem 2013;59(11):1583-94
- Tsui NB, Jiang P, Wong YF, et al. Maternal plasma RNA sequencing for genome-wide transcriptomic profiling and identification of pregnancy-associated transcripts. Clin Chem 2014;60(7):954-62
- Taglauer ES, Wilkins-Haug L and Bianchi DW. Review: Cell-free fetal DNA in the maternal circulation as an indicator of placental health and disease. Placenta 2014;35:S64-8
- Lo YMD. Non-invasive prenatal testing using massively parallel sequencing of maternal plasma DNA: from molecular karyotyping to fetal whole-genome sequencing. Reprod Biomed Online 2013;27(6):593-8