Abstract
The discovery of cell-free fetal DNA (cffDNA) in maternal plasma has enabled a paradigm shift in prenatal testing, allowing for safer, earlier detection of genetic conditions of the fetus. Non-invasive prenatal testing (NIPT) for fetal aneuploidies has provided an alternative, highly efficient approach to first-trimester aneuploidy screening, and since its inception has been rapidly adopted worldwide. Due to the genome-wide nature of some NIPT protocols, the commercial sector has widened the scope of cell-free DNA (cfDNA) screening to include sex chromosome aneuploidies, rare autosomal trisomies and sub-microscopic copy-number variants. These developments may be marketed as ‘expanded NIPT’ or ‘NIPT Plus’ and bring with them a plethora of ethical and practical considerations. Concurrently, cfDNA tests for single-gene disorders, termed non-invasive prenatal diagnosis (NIPD), have been developed for an increasing array of conditions but are less widely available. Despite the fact that all these tests utilise the same biomarker, cfDNA, there is considerable variation in key parameters such as sensitivity, specificity and positive predictive value depending on what the test is for. The distinction between diagnostics and screening has become blurred, and there is a clear need for the education of physicians and patients regarding the technical capabilities and limitations of these different forms of testing. Furthermore, there is a requirement for consistent guidelines that apply across health sectors, both public and commercial, to ensure that tests are validated and robust and that careful and appropriate pre-test and post-test counselling is provided by professionals who understand the tests offered.
Introduction
Cell-free fetal DNA
In pregnancy, DNA from the developing fetus can be detected in maternal plasma and is referred to as cell-free fetal DNA (cffDNA). The presence of cffDNA was proven during studies which detected Y chromosomal DNA within the plasma of women carrying male fetuses (Lo et al. 1997). cffDNA originates from the syncitiotrophoblast layer of the placenta (Flori et al. 2004, Alberry et al. 2007) and is released into the maternal bloodstream following endonuclease degradation as short dsDNA fragments with a median length of 143bp (Lo et al. 2010). cffDNA is therefore shorter on average than maternal cell-free DNA (cfDNA), which has a median length of 166bp, and derives from the natural lysis of cells from multiple bodily tissues, with the majority originating from haematopoietic cells. Intriguingly, cffDNA shows different fragment end sites to maternal cfDNA, with maternal cfDNA ends more commonly located within the linker regions between nucleosomes (Sun et al. 2018). The plasma of a pregnant woman therefore comprises a mixture of cfDNA from placental and maternal tissue, which can be used to test for genetic conditions in the fetus. Traditional invasive methods of collecting fetal genetic material for prenatal diagnosis via amniocentesis or chorionic villous sampling are associated with a small risk of miscarriage (Alfirevic et al. 2017), which can present a barrier to some parents, while no such risk is associated with the ‘non-invasive’ sampling of maternal peripheral blood. Although the perceived miscarriage risk is an important factor in shaping the views of patients, pregnant women also prefer non-invasive alternatives due to the pain and discomfort of invasive procedures and the ability to test earlier in the pregnancy (Hill et al. 2014). In addition to cffDNA, fetal cells isolated from the maternal bloodstream and cervix represent another source of fetal genetic material for non-invasive testing. The development of non-invasive testing methods using fetal cells has been reported for aneuploidy (Beaudet 2016) and microdeletions (Vossaert et al. 2018) but is not yet in clinical practice and hence is not discussed further in this review.
cffDNA is usually first detectable from 6 to 7 weeks gestation, with the earliest reported detection at 4.5 weeks (D’Aversa et al. 2018), often allowing non-invasive tests to be performed earlier in pregnancy than standard biochemical screens or invasive testing procedures (Fig. 1). The proportion of cffDNA to total cfDNA is referred to as the fetal fraction, which increases throughout pregnancy and can be as high as 30% in the third trimester. Following birth and the removal of the placenta, cffDNA is cleared from the maternal circulation within hours (Lo et al. 1999). Consequently, cffDNA is a suitable source of fetal genetic material, as it is specific to the ongoing pregnancy at the time of sampling. Since its discovery, cffDNA has revolutionised prenatal genetic testing, allowing the development of non-invasive screening methods for common aneuploidies, referred to as Non-invasive Prenatal Testing (NIPT) or Non-invasive Prenatal Screening (NIPS), and diagnostic testing for single-gene disorders, referred to as Non-Invasive Prenatal Diagnosis (NIPD). NIPT is a screening test, with positive results requiring confirmation via invasive testing. This is because there are several factors, such as confined placental mosaicism (CPM), which may lead to a false-positive NIPT result. On the other hand, CPM for single-gene disorders has not been reported. Consequently, cfDNA test results for single-gene disorders do not require confirmation via invasive methods and are therefore considered diagnostic. The characteristics of cffDNA that allow NIPT for aneuploidy and diagnosis of monogenic disorders are summarised in Table 1. This review will summarise the technical parameters, clinical utility and limitations of NIPT and NIPD.
This figure shows the timing of non-invasive prenatal tests within pregnancy relative to routine ultrasound scanning, biochemical screening for common aneuploidies and traditional methods of invasive testing. This diagram is intended to be illustrative rather than definitive and is based on current practices within the United Kingdom National Health Service. The timings of non-invasive tests offered by commercial providers may differ from those quoted here. First- and second-trimester biochemical screening are indicated by the ‘comb’ (combined screen: PAPP-A and free β-hCG) and ‘quadruple’ (AFP, total hCG, uE3 and inhibin A) test boxes. AS, anomaly scan; CVS, chorionic villous sample; NT, nuchal translucency; RDHO: relative haplotype dosage.
Citation: Reproduction 160, 5; 10.1530/REP-19-0591
Characteristics of cffDNA and the impact on prenatal screening for aneuploidy and diagnosis of monogenic conditions. A summary of how the provision of NIPT and NIPD are impacted by different characteristics of cffDNA.
Characteristics of cffDNA | Impact on NIPT for aneuploidy | Impact on NIPD for monogenic conditions |
---|---|---|
cffDNA is present in maternal plasma from early gestation | Early screening | Early diagnosis without risk of miscarriage |
cffDNA originates from the placenta | False-negative and false-positive results due to confined placental mosaicism (CPM) | No impact as CPM not reported for monogenic conditions |
The majority of cfDNA in maternal plasma originates from maternal tissues | Incidental detection of maternal chromosomal rearrangements including microdeletion and duplication syndromes, chromosomal mosaicism, sex chromosome aneuploidy and malignancy | Maternal somatic mosaicism must be excluded to avoid false positives by analysing maternal genomic DNA in parallel with cfDNA |
The relative proportion of cffDNA (fetal fraction) increases with gestational age | Ultrasound dating of pregnancy required Fetal fraction should be determined when testing to ensure sufficient cffDNA present |
Ultrasound dating of pregnancy required Fetal fraction should be determined when testing to ensure sufficient cffDNA present |
The placenta can shed fetal DNA into the maternal circulation for up to 6 weeks after demise of the fetal pole | To avoid discordant results from a vanishing twin, careful ultrasound is required | To avoid discordant results from a vanishing twin, careful ultrasound is required |
cffDNA is cleared from maternal circulation within hours of birth | Testing is pregnancy specific | Testing is pregnancy specific |
Non-invasive prenatal testing
Trisomies 13, 18 and 21
The first reports of NIPT for trisomies 13, 18 and 21 were published over a decade ago (Tsui et al. 2005, Lo et al. 2007, Chiu et al. 2008, Fan et al. 2008), and since then aneuploidy screening by NIPT has become firmly established within antenatal care pathways in many countries (Minear et al. 2015). The principle of NIPT lies in detecting a statistically significant increase in the relative dosage of chromosomal material in maternal cfDNA, which is attributed to the presence of a trisomic cell line in the fetus (Fig. 2). This is achieved using next generation sequencing (NGS) or microarray hybridisation of cfDNA in maternal plasma. The massively parallel functionalities of NGS and array technologies have enabled high-throughput testing on a scale amenable to population screening and the reliable determination of the fetal fraction via detection of paternally inherited single nucleotide polymorphisms (SNPs). Different methodologies may apply a targeted approach, in which only sequencing reads for defined chromosomes are generated. Alternatively, a genome-wide approach may be used in which sequencing reads are generated for all chromosomes, while analysis is restricted to the dosage chromosomes 13, 18 and 21. Targeted approaches are economically advantageous, while genome-wide approaches offer the potential to expand the scope of testing to include sex chromosome aneuploidies, rare autosomal trisomies and copy-number variants (covered in later sections of this review) without altering the underlying method.
This figure illustrates the concept of NIPT for aneuploidy. The cfDNA in the plasma of a pregnant woman is a mix of maternal cfDNA (black) and fetal (cffDNA) released from the fetal placenta (red). Measurement of cfDNA by NGS or microarray analysis is used to calculate the dosage of each chromosome. The maternal cfDNA and cffDNA are not distinguishable from each other but are measured in aggregate. An over-representation of sequences mapped to a particular chromosome compared to a reference chromosome indicates a fetal trisomy for that chromosome. Figure images were sourced from https://www.flaticon.com/.
Citation: Reproduction 160, 5; 10.1530/REP-19-0591
While not diagnostic, NIPT has been shown to have a much higher specificity and sensitivity than first-trimester biochemical screening and nuchal translucency measurement (Norton et al. 2015). Several studies have examined the testing parameters of NIPT, with a meta-analysis reporting the specificity for all three trisomies to be 99.87% and the sensitivity for trisomy 21 to be 99.7%, compared to 97.9% for trisomy 18 and 99.0% for trisomy 13 (Gil et al. 2017) . The lower sensitivity of NIPT for trisomies 13 and 18 is a result of the low average content of guanine and cytosine bases of these chromosomes compared to chromosome 21, which introduces non-uniform bias into sequencing reactions. NIPT therefore has a considerable advantage over first and second-trimester biochemical screening, as it has a lower rate of false positives, meaning fewer unnecessary invasive tests are offered in healthy pregnancies (Norton et al. 2015).
One key parameter to note is that while sensitivity and specificity are high, the positive predictive value varies both with prior risk factors, such as maternal age, and the individual trisomies (Petersen et al. 2017). NIPT has been implemented into healthcare systems worldwide, either as a contingent test for women at a defined risk level following first-trimester or serum screening, such as in Australia (Hui et al. 2017a), or replacing first-trimester biochemical screening entirely, such as in the Netherlands (van der Meij et al. 2019).
Limitations and quality control
NIPT is a highly accurate test when used for screening purposes; however, there are multiple limitations, which mean that it cannot be considered diagnostic. Consequently, robust quality assessment is required to ensure that minimum standards of testing and reporting are upheld between laboratories (Deans et al. 2019). For example, fetal fraction measurement is a key analytical and quality-control metric. Low fetal fraction has been linked to very early gestations, high maternal BMI, maternal medications, smoking and factors which lead to a smaller placenta, such as trisomies 13 and 18 (Kuhlmann-Capek et al. 2019). A fetal fraction quality threshold of 4% is commonly applied, below which results are often reported as inconclusive. Most NIPT platforms screen for the common trisomies with or without sex chromosome anomalies, and, while their use in routine screening for these trisomies seems clear, most other chromosomal rearrangements will not be detected. Thus, in the presence of fetal structural abnormalities on ultrasound, NIPT for the common aneuploidies should not be the test of choice, as there is a higher incidence of chromosomal rearrangements in this pregnancy cohort (Al Toukhi et al. 2019).
There are several potential causes of a discordant NIPT result. These include a ‘vanishing twin’: an aneuploid twin pregnancy that spontaneously miscarries early in pregnancy but still releases cffDNA into the maternal blood (Alberry et al. 2007). In this scenario, the cffDNA released by the placenta after the demise of the aneuploid fetus may be detected by early NIPT and falsely attributed to a euploid twin. As NIPT analyses all cfDNA, both fetal and maternal, in maternal plasma, detection of abnormal maternal cell lines is another potential aetiology of discordant results. These include maternal cytogenetic anomalies, either in constitutional or mosaic form, but also malignancies where ‘chaotic’ results may indicate circulating cell-free tumour DNA (Bianchi et al. 2015). Indeed, women with known malignancies should not have NIPT, as the results cannot be accurately interpreted (Lenaerts et al. 2019). Finally, as mentioned previously, cell-free ‘fetal’ DNA may not represent the genotype of the fetus: there is potential for NIPT to produce inaccurate results either due to CPM for aneuploidy (Pan et al. 2013) or due to complete discordance between fetal and placental genotypes (Verweij et al. 2014). This can result in both discordant positive and negative results depending on the predominant cell line in the placenta (Hartwig et al. 2017).
Sex chromosome aneuploidies
Sex chromosome aneuploidies, such as Turner syndrome (45,X) and Klinefelter syndrome (47,XXY), are variably reported using NIPT, for example, they are not reported in the Netherlands (van der Meij et al. 2019). The sensitivity of NIPT for sex chromosome aneuploidies is lower than for the common trisomies, with much lower positive predictive values (PPVs) than NIPT for Down syndrome, particularly for Turner syndrome. False-positive rates of up to 90% have been reported in low-risk cohorts, which raises questions about the clinical utility of this information (Reiss et al. 2017), although lower false-positive rates are reported for cohorts with ultrasound anomalies such as cystic hygroma. The reasons for such high false-positive rates for sex chromosome aneuploidies include CPM, but also constitutional or mosaic sex chromosome aneuploidies in the mother, such as X chromosome segmental duplications, triple X syndrome and mosaic Turner syndrome. In a recent study confirming the poor PPV, 20% of false‐positive sex chromosome aneuploidy results were due to a maternal aneuploidy and a further 23% of fetal X chromosome copy-number variants were maternally inherited (Zhang et al. 2019a). Such results highlight the variability in clinical presentation of these conditions, as they may be detected incidentally in pregnant women with no apparent clinical features. The rationale for including sex chromosome aneuploidies in screening programmes is therefore debated, as ‘affected’ individuals may not have significant adverse health outcomes. However, a potential advantage of NIPT for sex chromosome aneuploidies is that detection during pregnancy may allow early initiation of postnatal interventions that may improve neurodevelopmental outcomes.
Rare autosomal trisomies
The term ‘rare autosomal trisomy’ (RAT) refers to a trisomy for any autosome other than 13, 18 and 21. Constitutional forms of these aneuploidies are almost invariably lethal, and hence the overwhelming majority of cases represent mosaicism which may be confined to placental tissue (Grati et al. 2019). Originally, NIPT methods were designed solely to detect trisomies 13, 18 and 21. However, NGS-based NIPT methods generate low-depth sequencing coverage for all autosomes, which has allowed retrospective re-analysis of these datasets to detect trisomies for any chromosome (Pertile et al. 2017). Several studies applying this analytical methodology have revealed that the prevalence of RATs is approximately 0.1–0.3% in general obstetric population cohorts (Table 2). Trisomy 7 is the most commonly detected RAT, while trisomies 15, 16 and 22 are more frequently detected via NIPT than previous studies using chorionic villous sampling data (Benn et al. 2019) (Fig. 3). Constitutional RATs are usually associated with spontaneous miscarriage, but mosaic RATs may be associated with a range of adverse outcomes such as placental insufficiency, low birth weight, miscarriage and structural anomalies due to fetal mosaicism (Scott et al. 2018). In addition, CPM for a chromosome containing imprinted regions can lead to a clinical phenotype via generation of uniparental disomy in the developing fetus following trisomy rescue. A key example is uniparental disomy for chromosome 15, which causes Prader–Willi syndrome or Angelman syndrome, dependent on a maternal or paternal origin, respectively. However, mosaic RATs are also associated with normal births: a recent meta-analysis reported that 41% of RATs detected via NIPT resulted in a normal postnatal outcome (Benn et al. 2019). Thus, while extending the diagnostic scope of NGS-based NIPT to include detection of RATs is possible without significant amendment of most technical laboratory protocols, the utility of this approach is controversial and there is as yet no consensus on value from the clinical community.
This figure shows the reported prevalence of each RAT from the studies shown in Table 2 as a proportion of the total cases tested (n = 157,159). The absolute numbers of each RAT are indicated above each column.
Citation: Reproduction 160, 5; 10.1530/REP-19-0591
Studies of RAT detection using NIPT. This table summarises the results of published studies reporting rare autosomal trisomies (RATs) detected via NIPT. This table only includes studies publishing the prevalence of individual trisomies in each cohort.
Study | Population | Study type | Study size | Total RATs |
---|---|---|---|---|
Lau et al. (2014) | General population | Prospective | 1,982 | 7 (0.35%) |
Brady et al. (2016) | Increased risk | Prospective | 4,000 | 11 (0.28%) |
Fiorentino et al. (2017) | Increased risk | Prospective | 12,078 | 17 (0.14%) |
Pertile et al. (2017) | General population | Retrospective | 89,817 | 306 (0.34%) |
Pescia et al. (2017) | Not specified | Prospective | 6,388 | 50 (0.78%) |
Scott et al. (2018) | General population | Prospective | 23,388 | 28 (0.12%) |
Wan et al. (2018) | General population | Retrospective | 15,362 | 53 (0.35%) |
Van Opstal et al. (2018) | Increased risk | Prospective | 2,527 | 29 (0.91%) |
Chatron et al. (2019 | Increased risk | Prospective | 1,617 | 10 (0.62%) |
Copy-number variants
NIPT has also been extended to the detection of chromosomal deletions and duplications within the fetal genome, by applying the same principles of dosage as for aneuploidy analysis (Advani et al. 2017). Microdeletions and microduplications are copy-number variants (CNVs) which lie below the resolution of traditional karyotyping methods and are associated with a broad range of genetic syndromes. While individually rare, these conditions are collectively common and do not exhibit a maternal age affect, unlike the common trisomies. Pathogenic CNVs can occur across the genome but around 25% are recurrent, the most common being the 22q11.2 deletion, which is causative of Di-George syndrome and has been demonstrated to have a prevalence of 1 in 992 in a low-risk obstetric population (Grati et al. 2015). While NIPT can be extended to include CNV screening, the majority of commercial platforms only report the detection of several recurrent microdeletion syndromes (Table 3), with only one claiming to detect all CNVs that are 7 Mb or greater. However, as most microdeletion syndromes and non-recurrent pathogenic CNVs are smaller than 5 Mb, such strategies will only detect the minority of relevant CNVs. Those platforms targeting specific recurrent microdeletions are also limited, as non-recurrent CNVs occur across the genome. In a review of prenatal cases analysed in our Regional Cytogenetic Laboratory from 1997 to 2013, 173 pathogenic CNVs were detected in 23,000 cases, out of which 77% were non-recurrent and would not be detected by the currently available commercial platforms (Chitty et al. 2018). Not only is the sensitivity poor for most of these conditions, but the PPVs are considerably lower than for the common trisomies and can vary significantly depending on the patient’s clinical details. Using the 22q11.2 deletion as an example, the PPV of NIPT can range from 21% in low-risk pregnancies (Petersen et al. 2017) to 50–97% in pregnancies with ultrasound anomalies (Helgeson et al. 2015, Gross et al. 2016). It is clear that practitioners offering extended NIPT, which include CNVs, should provide comprehensive counselling before and after testing, including the possibility of no findings and the need for confirmation of positive results with invasive testing (Grati & Gross 2019). There is also the consideration that many CNV syndromes present with variable expression, and accurate prediction of phenotypic severity in the absence of ultrasound findings is not possible. For the reasons discussed here, NIPT for CNVs and RATs is not currently endorsed by any professional society, and some national bodies do not endorse its use for fetal sex determination in the absence of a family history of sex-linked disorders.
Recurrent chromosomal deletions causing genetic syndromes commonly offered in expanded NIPT by commercial providers.
Chromosomal locus | Condition |
---|---|
1p36 | 1p36 deletion syndrome |
4p16 | Wolf–Hirschorn syndrome |
5p | Cri du Chat syndrome |
8q24 | 8q24 deletion syndrome |
11q23 | Jacobsen syndrome |
15q11–13 | Angelman syndrome and Prader–Willi syndrome |
22q11.2 | Di-George syndrome |
Patient and health professional perspectives and ethical issues
Uptake of NIPT has been high with both women and health professionals welcoming the potential for an earlier and more accurate screening test, which can result in increased detection of Down syndrome with a significantly reduced rate of invasive testing (Chitty et al. 2016). However, the simplicity of sample collection and the number of routine blood tests performed during pregnancy could mean that women do not fully consider the consequences of a ‘high chance’ result without appropriate counselling. This has led to calls for better patient and health professional education to ensure that women have the opportunity to make informed choices regarding testing (Lewis et al. 2017). Despite fears that NIPT would increase termination rates of fetuses with Down syndrome, data do not support this conclusion. Instead, findings from international studies suggest that many women take this more accurate screening test to gain information about their baby rather than to terminate a pregnancy (Hill et al. 2017). Another consideration is the potential for NIPT to facilitate sex-selective termination of pregnancy, as it is available very early in pregnancy and can be used to determine genetic sex. At present, ultrasound remains the primary method for prenatal determination of fetal sex in this context, and evidence that NIPT can facilitate sex selection is anecdotal (Bowman-Smart et al. 2019).
NIPD for single-gene disorders
The expansion of diagnostic testing for single-gene disorders using cfDNA has been comparatively slower than NIPT for aneuploidy screening. This is probably due to multiple factors, including the relative rarity of some conditions, the technical complexity of testing and the need for development on an individual family basis in many instances, meaning that, unlike NIPT for Down syndrome, there has been less potential for commercialisation (Jenkins et al. 2018).
Methodologies for NIPD can be broadly divided into two categories. First, there is the detection of a genetic variant in the fetus that is not present in the mother. This approach is appropriate for the detection of paternally inherited variants for dominant and recessive conditions and for variants that have arisen de novo at conception. Secondly, there are dosage-based techniques targeting genetic variants which are carried by the mother and are therefore present in maternal cfDNA. Detection of dosage imbalances of these variants in the total circulating cfDNA of pregnant women can be used to predict the fetal genotype (Fig. 4).
NIPD via low level variant detection (A) and relative mutation dosage (B). The chromosome ideograms show the mutation and WT allele in red and blue, respectively. (A) In situations where the mother is not a carrier of the variant of interest, the presence or absence of the mutation at low levels within maternal plasma can be used for diagnosis in the fetus. This applies in cases of a dominant paternal condition (such as achondroplasia), chromosomal sex (using Y chromosome markers) or for recessive conditions in which the father and mother are heterozygous carriers of different mutations. (B) This illustrates relative mutation dosage in an autosomal recessive disease model. When both parents are carriers of the same mutation, the dosage of the mutant and WT alleles in maternal plasma can be used to infer the fetal genotype.
Citation: Reproduction 160, 5; 10.1530/REP-19-0591
De novo and paternally inherited conditions
The first clinical use of NIPD for the detection of paternally inherited markers was for fetal sex determination using quantitative PCR for Y chromosome sequences (Devaney et al. 2011) and for the detection of Rhesus D positive (RHD+) fetuses in Rhesus D negative (RHD-) mothers (Finning & Chitty 2008). Non-invasive fetal sex determination is now widely used across Europe to direct invasive testing in pregnancies at risk of sex-linked conditions where it has been shown to be highly accurate, cost effective and reduces invasive testing by around 50% (Hill et al. 2011). NIPD for fetal sex determination can also clarify the genetic sex of the fetus when ambiguous genitalia are detected via ultrasound, which informs parental counselling. Fetal RHD typing was initially used to direct fetal monitoring and treatment in pregnancies at high risk of haemolytic disease of the new-born (Finning & Chitty 2008). However, in many parts of Europe this is now used to direct routine immunoglobulin therapy in RHD- mothers (Clausen et al. 2019), but the clinical and cost effectiveness in some parts of the world has been challenged (Moise et al. 2019).
Subsequently NIPD methods for detecting de novo and paternally inherited pathogenic variants were developed, since these can easily be distinguished against the high background of maternal cfDNA. This has been successfully developed for autosomal dominant disorders such as the FGFR3-related skeletal dysplasias: first, via restriction enzyme digest to target individual mutations (Chitty et al. 2011) and then extended to targeted NGS panels allowing multiple variants to be assessed in a single and more accurate test (Chitty et al. 2015). Application of NIPD to this patient cohort is strengthened by well-characterised fetal phenotypes on ultrasound scanning. For autosomal recessive conditions such as cystic fibrosis (Hill et al. 2015) and β-thalassaemia (Xiong et al. 2015), where the father and mother are heterozygous for different pathogenic variants, targeted testing for the paternal variant in the cfDNA can be performed to offer paternal exclusion testing. Invasive testing will then only be required if the paternal mutant allele is detected.
Bespoke amplicon-based NGS assays can also be developed for a range of rare monogenic diseases caused by known mutations specific to a particular family. As each assay is developed and validated on an individual family basis, bespoke testing is considerably more expensive than invasive testing and other forms of NIPD (Verhoef et al. 2016). Some have argued that, as the recurrence risk in these situations is extremely low, it may not be appropriate to offer this testing within a publicly funded healthcare system (Wilkie & Goriely 2017).
X-linked and recessive conditions
Development of NIPD is more technically challenging for X-linked conditions and for autosomal recessive conditions when both parents are carriers of the same mutation. This is due to the high background of the relevant mutation from maternal tissue in the circulating cfDNA.
Relative mutation dosage
NIPD using relative mutation dosage (RMD) requires the precise quantification of mutant and WT alleles in cfDNA, and the application of statistical methods to clarify that measured imbalances reflect the signal of the fetal genotype rather than technical noise. Unfortunately, standard protocols of NGS are insufficiently sensitive for these applications, as amplification bias between mutant and WT alleles can lead to inaccurate allelic fractions. Potential solutions to these challenges include the use of nested PCR (Xiong et al. 2018, Cutts et al. 2019), unique molecular indexes and synthetic reference amplicons that have known amplification dynamics (Tsao et al. 2019).
Another key technique in this area is digital PCR (dPCR). dPCR is a highly sensitive technique in which a single PCR reaction is separated into many thousands of partitions. Detection of the presence or absence of an allele-specific fluorescent signal from each partition allows the concentration of the target sequence to be precisely quantified according to Poisson statistics. Proof-of-principle studies for NIPD using dPCR have been reported for several recessive and X-linked conditions, including β-thalassemia (Lun et al. 2008, Camunas-Soler et al. 2018), sickle cell disease (Barrett et al. 2012), haemophilia (Tsui et al. 2011, Hudecova et al. 2017) and recessive forms of deafness (Chang et al. 2016). While dPCR is highly sensitive, it has limited capacity for multiplexing comparative to NGS, which restricts the number of mutations that can be assayed in one test.
Relative haplotype dosage
Rather than directly detecting pathogenic variants, NIPD using relative haplotype dosage (RHDO) determines which parental haplotypes have been inherited by the fetus based on the inheritance of SNPs at the locus of the relevant disease gene (Lo et al. 2010). Using NGS, the inherited paternal haplotype can be determined by detection of low-level SNPs in the cfDNA which differ from the maternal haplotype, while the inherited maternal haplotype can be determined by changes in dosage of SNPs which differ from the paternal haplotype. The inherited haplotypes are then compared to those inherited by a previous pregnancy, usually an affected proband, to determine the genetic status of the fetus (Fig. 5). In this manner, RHDO employs both low-level variant detection and dosage techniques to determine haplotype inheritance. As multiple SNPs are used for classifying haplotypes, RHDO is not affected by the technical noise of NGS to the same degree as RMD, and hence standard NGS protocols are sufficient without modifications for molecular counting.
(A) This figure illustrates the method of RHDO using an example pedigree for cystic fibrosis, in which the parents are both heterozygous for the common CFTR c.1521_1523del p.Phe508del mutation (F508Δ) and have a daughter affected with cystic fibrosis. (B) Sequencing of genomic DNA from the mother, father and daughter allows delineation of the parental haplotypes associated with each mutant and WT allele. These haplotypes are defined by informative heterozygous SNPs, indicated by the dark and light blue (paternal) and orange (maternal) boxes, that lie within and surrounding the CFTR gene. A proband sample is required for this, as NGS cannot determine haplotypes using only parental samples: the short read length prevents phasing a particular SNP onto the same chromosome as the mutation. (C) Sequencing of cfDNA from maternal plasma then allows the haplotypes inherited by the fetus to be detected through dosage imbalance of the maternal haplotypes and low-level detection of the paternal haplotype. The four different fetal haplotypes are shown, along with the maternal haplotype background, and the resulting relative dosage of each haplotype detected in the cfDNA. WT: wild type.
Citation: Reproduction 160, 5; 10.1530/REP-19-0591
Notably, RHDO is able to determine the inheritance of complex genomic variants which are beyond the resolution of cfDNA fragmentation, such as exonic deletions and the intron-22-related inversions within the F8 gene which cause severe haemophilia A (Hudecova et al. 2017). RHDO can also be applied to genes with homologous pseudogenes that complicate direct mutation detection, most notably CYP21A2-related congenital adrenal hyperplasia (New et al. 2014).
In contrast to RMD approaches, RHDO has been implemented clinically, and services for Duchenne muscular dystrophy (Parks et al. 2016), spinal muscular atrophy (Parks et al. 2017) and cystic fibrosis (Chandler et al. 2019) are now available in the United Kingdom National Health Service. Over 90 cases of proband-based RHDO have been reported for several monogenic disorders, with no false-positive or false-negative results (Table 4). The limitations of this approach include the high cost of testing and the potential for inconclusive results due to recombination events within the target locus (Chandler et al. 2019). Moreover, RHDO may not be applicable in pregnancies with consanguineous parents, as the technique relies on a large number of different SNPs to differentiate between maternal and paternal haplotypes.
Studies reporting NIPD using relative haplotype dosage for a range of monogenic conditions.
Publication | Condition | Cases | Methodology | Sensitivity |
---|---|---|---|---|
Lo et al. (2010) | BT | 1 | Proband based | 100% |
New et al. (2014) | CAH | 14 | Proband based | 100% |
Parks et al. (2016) | DMD and BMD | 9 | Proband based | 100% |
Parks et al. (2017) | SMA | 16 | Proband based | 100% |
Hui et al. (2017b) | CAH, BT, EVCS, F8-H and HS | 13 | Parental - linked read | 100% |
Hudecova et al. (2017) | F8-H | 3 | Proband-based | 100% |
Vermeulen et al. (2017) | CF, CAH and BT | 18 | Parental - TLA | 100% |
Jang et al. (2018) | DMD | 5 | Parental - linked read | 100% |
Chandler et al. (2019) | CF | 51 | Proband-based | 100% |
BMD, Becker muscular dystrophy; BT, β-thalassemia; CAH, congenital adrenal hyperplasia; CF, cystic fibrosis; DMD, Duchenne muscular dystrophy; EVCS, Ellis–van Creveld syndrome; F8-H, factor 8 haemophilia; HS, Hunter syndrome; SMA, spinal muscular atrophy; TLA, targeted locus amplification.
Currently, clinical RHDO services require familial samples for haplotype phasing, most commonly genomic DNA from the father, mother and an affected proband. This unfortunately restricts the application of RHDO to families with children or where DNA from a previous child or pregnancy has been stored. Proband-free RHDO, using methods to haplotype the parents directly, has so far been shown to be possible by two different approaches: targeted locus amplification (TLA) (Vermeulen et al. 2017) and microfluidics-based linked-read sequencing (Hui et al. 2017b , Jang et al. 2018). In addition, long-read sequencing technologies offer the potential to directly haplotype parental genomic DNA, although proof-of-principle is yet to be reported for this approach.
Commercial NIPD for screening low-risk pregnancies
The development of NIPD has thus far been concentrated on pregnancies at high risk of single-gene disorders, either due to a pre-existing family history or ultrasound findings consistent with a specific condition. There are now, however, increasing efforts by the commercial sector to develop NIPD to screen the general population for monogenic disorders. Two key areas are emerging: low-level variant NIPD for de novo mutations in dominant disease genes, such as for Noonan syndrome and achondroplasia (Zhang et al. 2019b), and RMD approaches for mutations with high population carrier frequencies, such as sickle cell disease and spinal muscular atrophy, which are among the most common indications for invasive testing (Tsao et al. 2019). Both of these tests are now commercially available, and it is argued that these applications have considerable potential to impact prenatal care by providing definitive diagnosis of genetic conditions early in pregnancy and by facilitating the potential for postnatal or in utero treatment.
However, there remain many issues with the provision of these tests. In the rare disease area, the technical information on gene coverage and test sensitivity from commercial providers is limited and compounded by a lack of follow-up data for reported cases. The provision of these tests to women with no family history or clinical indication, such as ultrasound anomalies, may lead either to unnecessary stress and concern or inappropriate reassurance that the fetus does not have a genetic condition. In particular, mutation agnostic approaches may detect variants of uncertain clinical significance, which pose major counselling issues. Furthermore, as there is currently limited follow-up or validation data available, these tests should be used with caution and positive results confirmed by invasive testing. Larger scale studies with comprehensive follow-up are required to determine the true sensitivity and specificity of these tests. A significant concern is that of false-negative results, which can either be due to incomplete coverage of genes tested or other potential causes, such as variation in enzymatic cut-sites in the fragmented cfDNA (Sun et al. 2018). Conversely, false positives may occur if somatic mosaicism for a variant is misinterpreted as a fetal genotype, and consequently maternal genomic DNA should always be simultaneously tested in order to exclude this possibility. These concerns mean that rigorous standards of technical validation should be applied to all new tests whether developed in the commercial or public health sector and that parental counselling should include all the potential technical limitations.
Conclusions
Non-invasive prenatal tests based on analysis of cfDNA have transformed prenatal care. NIPT provides a cost-effective, high-sensitivity screening test for the common trisomies, and its global implementation has dramatically reduced the number of invasive prenatal procedures performed. Conversely, NIPD for single-gene disorders is less widely available and may be significantly more expensive dependent on the approach chosen. Given the high cost of particular NIPD methods, such as bespoke mutation exclusion and RHDO, a wider debate is required on who should be offered testing and for which conditions within publicly funded healthcare systems. The entire fetal genome is represented in cffDNA, and genome-wide sequencing methodologies have allowed commercial providers to report on a broader range of fetal genetic abnormalities, including sex chromosome abnormalities, RATs and CNV syndromes. However, these developments are controversial, and the low PPVs, debatable clinical utility and associated counselling challenges mean that screening for these conditions is not currently supported by any international society. In addition, there is now commercial interest in providing NIPD to screen for monogenic conditions in low-risk pregnancies. The continuing education of physicians and patients about the technical capabilities and limitations of different testing methods is crucial to ensure these tests are implemented appropriately to provide maximal benefit for families.
Declaration of interest
L S C has had academic grants from the NIHR, Action Medical Research and GOSH Children’s Charity for the development of NIPD. The other authors have nothing to disclose.
Funding
J S and E S are fully funded and L S C and N C are partially funded by the NIHR Biomedical Research Centre at Great Ormond Street Hospital. The views expressed are those of the authors and not necessarily those of the NHS, the NIHR or the Department of Health.
Author contribution statement
J S and L S C wrote the manuscript, with editing performed by E S and N C. J S designed the tables and figures, excluding Table 1 which was designed by LSC.
References
Advani HV, Barrett AN, Evans MI & Choolani M 2017 Challenges in non-invasive prenatal screening for sub-chromosomal copy number variations using cell-free DNA. Prenatal Diagnosis 37 1067–1075. (https://doi.org/10.1002/pd.5161)
Al Toukhi S, Chitayat D, Keunen J, Roifman M, Seaward G, Windrim R, Ryan G & Van Mieghem T 2019 Impact of introduction of noninvasive prenatal testing on uptake of genetic testing in fetuses with central nervous system anomalies. Prenatal Diagnosis 39 544–548. (https://doi.org/10.1002/pd.5466)
Alberry M, Maddocks D, Jones M, Abdel Hadi M, Abdel-Fattah S, Avent N & Soothill PW 2007 Free fetal DNA in maternal plasma in anembryonic pregnancies: confirmation that the origin is the trophoblast. Prenatal Diagnosis 27 415–418. (https://doi.org/10.1002/pd.1700)
Alfirevic Z, Navaratnam K & Mujezinovic F 2017 Amniocentesis and chorionic villus sampling for prenatal diagnosis. Cochrane Database of Systematic Reviews 9 Cd003252. (https://doi.org/10.1002/14651858.CD003252.pub2)
Barrett AN, Mcdonnell TC, Chan KC & Chitty LS 2012 Digital PCR analysis of maternal plasma for noninvasive detection of sickle cell anemia. Clinical Chemistry 58 1026–1032. (https://doi.org/10.1373/clinchem.2011.178939)
Beaudet AL 2016 Using fetal cells for prenatal diagnosis: history and recent progress. American Journal of Medical Genetics: Part C, Seminars in Medical Genetics 172 123–127. (https://doi.org/10.1002/ajmg.c.31487)
Benn P, Malvestiti F, Grimi B, Maggi F, Simoni G & Grati FR 2019 Rare autosomal trisomies: comparison of detection through cell-free DNA analysis and direct chromosome preparation of chorionic villus samples. Ultrasound in Obstetrics and Gynecology 54 458–467. (https://doi.org/10.1002/uog.20383)
Bianchi DW, Chudova D, Sehnert AJ, Bhatt S, Murray K, Prosen TL, Garber JE, Wilkins-Haug L, Vora NL & Warsof S et al. 2015 Noninvasive prenatal testing and incidental detection of occult maternal malignancies. JAMA 314 162–169. (https://doi.org/10.1001/jama.2015.7120)
Bowman-Smart H, Savulescu J, Gyngell C, Mand C & Delatycki MB 2019 Sex selection and non-invasive prenatal testing: a review of current practices, evidence, and ethical issues. Prenatal Diagnosis In press. (https://doi.org/10.1002/pd.5555)
Brady P, Brison N, Van Den Bogaert K, De Ravel T, Peeters H, Van Esch H, Devriendt K, Legius E & Vermeesch JR 2016 Clinical implementation of NIPT - technical and biological challenges. Clinical Genetics 89 523-530. (https://doi.org/10.1111/cge.12598)
Camunas-Soler J, Lee H, Hudgins L, Hintz SR, Blumenfeld YJ, El-Sayed YY & Quake SR 2018 Noninvasive prenatal diagnosis of single-gene disorders by use of droplet digital PCR. Clinical Chemistry 64 336–345. (https://doi.org/10.1373/clinchem.2017.278101)
Chandler NJ, Ahlfors H, Drury S, Mellis R, Hill M, Mckay FJ, Collinson C, Hayward J, Jenkins L & Chitty LS 2019 Noninvasive prenatal diagnosis for cystic fibrosis: implementation, uptake, outcome, and implications. Clinical Chemistry 66 207-216. (https://doi.org/10.1373/clinchem.2019.305011)
Chang MY, Kim AR, Kim MY, Kim S, Yoon J, Han JJ, Ahn S, Kang C & Choi BY 2016 Development of novel noninvasive prenatal testing protocol for whole autosomal recessive disease using picodroplet digital PCR. Scientific Reports 6 37153. (https://doi.org/10.1038/srep37153)
Chatron N, Till M, Abel C, Bardel C, Ramond F, Sanlaville D & Schluth-Bolard C 2019 Detection of rare autosomal trisomies through non-invasive prenatal testing: benefits for pregnancy management. Ultrasound in Obstetrics & Gynecology 53 129-130. (https://doi.org/10.1002/uog.20094)
Chitty LS, Griffin DR, Meaney C, Barrett A, Khalil A, Pajkrt E & Cole TJ 2011 New aids for the non-invasive prenatal diagnosis of achondroplasia: dysmorphic features, charts of fetal size and molecular confirmation using cell-free fetal DNA in maternal plasma. Ultrasound in Obstetrics and Gynecology 37 283–289. (https://doi.org/10.1002/uog.8893)
Chitty LS, Mason S, Barrett AN, Mckay F, Lench N, Daley R & Jenkins LA 2015 Non-invasive prenatal diagnosis of achondroplasia and thanatophoric dysplasia: next-generation sequencing allows for a safer, more accurate, and comprehensive approach. Prenatal Diagnosis 35 656–662. (https://doi.org/10.1002/pd.4583)
Chitty LS, Wright D, Hill M, Verhoef TI, Daley R, Lewis C, Mason S, Mckay F, Jenkins L & Howarth A et al. 2016 Uptake, outcomes, and costs of implementing non-invasive prenatal testing for Down’s syndrome into NHS maternity care: prospective cohort study in eight diverse maternity units. BMJ 354 i3426. (https://doi.org/10.1136/bmj.i3426)
Chitty LS, Hudgins L & Norton ME 2018 Current controversies in prenatal diagnosis 2: cell-free DNA prenatal screening should be used to identify all chromosome abnormalities. Prenatal Diagnosis 38 160–165. (https://doi.org/10.1002/pd.5216)
Chiu RW, Chan KC, Gao Y, Lau VY, Zheng W, Leung TY, Foo CH, Xie B, Tsui NB & Lun FM et al. 2008 Noninvasive prenatal diagnosis of fetal chromosomal aneuploidy by massively parallel genomic sequencing of DNA in maternal plasma. PNAS 105 20458–20463. (https://doi.org/10.1073/pnas.0810641105)
Clausen FB, Rieneck K, Krog GR, Bundgaard BS & Dziegiel MH 2019 Noninvasive antenatal screening for fetal RHD in RhD negative women to guide targeted anti-D prophylaxis. Methods in Molecular Biology 1885 347–359. (https://doi.org/10.1007/978-1-4939-8889-1_23)
Cutts A, Vavoulis DV, Petrou M, Smith F, Clark B, Henderson S & Schuh A 2019 A method for noninvasive prenatal diagnosis of monogenic autosomal recessive disorders. Blood 134 1190–1193. (https://doi.org/10.1182/blood.2019002099)
D’aversa E, Breveglieri G, Pellegatti P, Guerra G, Gambari R & Borgatti M 2018 Non-invasive fetal sex diagnosis in plasma of early weeks pregnants using droplet digital PCR. Molecular Medicine 24 14. (https://doi.org/10.1186/s10020-018-0016-7)
Deans ZC, Allen S, Jenkins L, Khawaja F, Gutowska-Ding W, Patton SJ, Chitty LS & Hastings RJ 2019 Ensuring high standards for the delivery of NIPT world-wide: development of an international external quality assessment scheme. Prenatal Diagnosis 39 379–387. (https://doi.org/10.1002/pd.5438)
Devaney SA, Palomaki GE, Scott JA & Bianchi DW 2011 Noninvasive fetal sex determination using cell-free fetal DNA: a systematic review and meta-analysis. JAMA 306 627–636. (https://doi.org/10.1001/jama.2011.1114)
Fan HC, Blumenfeld YJ, Chitkara U, Hudgins L & Quake SR 2008 Noninvasive diagnosis of fetal aneuploidy by shotgun sequencing DNA from maternal blood. PNAS 105 16266–16271. (https://doi.org/10.1073/pnas.0808319105)
Finning KM & Chitty LS 2008 Non-invasive fetal sex determination: impact on clinical practice. Seminars in Fetal and Neonatal Medicine 13 69–75. (https://doi.org/10.1016/j.siny.2007.12.007)
Fiorentino F, Bono S, Pizzuti F, Duca S, Polverari A, Faieta M, Baldi M, Diano L & Spinella F 2017 The clinical utility of genome-wide non invasive prenatal screening. Prenatal Diagnosis 37 593– 601. (https://doi.org/10.1002/pd.5053)
Flori E, Doray B, Gautier E, Kohler M, Ernault P, Flori J & Costa JM 2004 Circulating cell-free fetal DNA in maternal serum appears to originate from cyto- and syncytio-trophoblastic cells. Case report. Human Reproduction 19 723–724. (https://doi.org/10.1093/humrep/deh117)
Gil MM, Accurti V, Santacruz B, Plana MN & Nicolaides KH 2017 Analysis of cell-free DNA in maternal blood in screening for aneuploidies: updated meta-analysis. Ultrasound in Obstetrics and Gynecology 50 302–314. (https://doi.org/10.1002/uog.17484)
Grati FR & Gross SJ 2019 Noninvasive screening by cell-free DNA for 22q11.2 deletion: benefits, limitations, and challenges. Prenatal Diagnosis 39 70–80. (https://doi.org/10.1002/pd.5391)
Grati FR, Molina Gomes D, Ferreira JC, Dupont C, Alesi V, Gouas L, Horelli-Kuitunen N, Choy KW, Garcia-Herrero S & De La Vega AG et al. 2015 Prevalence of recurrent pathogenic microdeletions and microduplications in over 9500 pregnancies. Prenatal Diagnosis 35 801–809. (https://doi.org/10.1002/pd.4613)
Grati FR, Ferreira J, Benn P, Izzi C, Verdi F, Vercellotti E, Dalpiaz C, D'ajello P, Filippi E & Volpe N et al. 2019 Outcomes in pregnancies with a confined placental mosaicism and implications for prenatal screening using cell-free DNA. Genetics in Medicine 22 309–316. (https://doi.org/10.1038/s41436-019-0630-y)
Gross SJ, Stosic M, Mcdonald-Mcginn DM, Bassett AS, Norvez A, Dhamankar R, Kobara K, Kirkizlar E, Zimmermann B & Wayham N et al. 2016 Clinical experience with single-nucleotide polymorphism-based non-invasive prenatal screening for 22q11.2 deletion syndrome. Ultrasound in Obstetrics and Gynecology 47 177–183. (https://doi.org/10.1002/uog.15754)
Hartwig TS, Ambye L, Sorensen S & Jorgensen FS 2017 Discordant non-invasive prenatal testing (NIPT) – a systematic review. Prenatal Diagnosis 37 527–539. (https://doi.org/10.1002/pd.5049)
Helgeson J, Wardrop J, Boomer T, Almasri E, Paxton WB, Saldivar JS, Dharajiya N, Monroe TJ, Farkas DH & Grosu DS et al. 2015 Clinical outcome of subchromosomal events detected by whole-genome noninvasive prenatal testing. Prenatal Diagnosis 35 999–1004. (https://doi.org/10.1002/pd.4640)
Hill M, Finning K, Martin P, Hogg J, Meaney C, Norbury G, Daniels G & Chitty LS 2011 Non-invasive prenatal determination of fetal sex: translating research into clinical practice. Clinical Genetics 80 68–75. (https://doi.org/10.1111/j.1399-0004.2010.01533.x)
Hill M, Compton C, Karunaratna M, Lewis C & Chitty L 2014 Client views and attitudes to non-invasive prenatal diagnosis for sickle cell disease, thalassaemia and cystic fibrosis. Journal of Genetic Counseling 23 1012–1021. (https://doi.org/10.1007/s10897-014-9725-4)
Hill M, Twiss P, Verhoef TI, Drury S, Mckay F, Mason S, Jenkins L, Morris S & Chitty LS 2015 Non-invasive prenatal diagnosis for cystic fibrosis: detection of paternal mutations, exploration of patient preferences and cost analysis. Prenatal Diagnosis 35 950–958. (https://doi.org/10.1002/pd.4585)
Hill M, Barrett A, Choolani M, Lewis C, Fisher J & Chitty LS 2017 Has noninvasive prenatal testing impacted termination of pregnancy and live birth rates of infants with Down syndrome? Prenatal Diagnosis 37 1281–1290. (https://doi.org/10.1002/pd.5182)
Hudecova I, Jiang P, Davies J, Lo YMD, Kadir RA & Chiu RWK 2017 Noninvasive detection of F8 int22h-related inversions and sequence variants in maternal plasma of hemophilia carriers. Blood 130 340–347. (https://doi.org/10.1182/blood-2016-12-755017)
Hui L, Hutchinson B, Poulton A & Halliday J 2017a Population-based impact of noninvasive prenatal screening on screening and diagnostic testing for fetal aneuploidy. Genetics in Medicine 19 1338–1345. (https://doi.org/10.1038/gim.2017.55)
Hui WW, Jiang P, Tong YK, Lee WS, Cheng YK, New MI, Kadir RA, Chan KC, Leung TY & Lo YM et al. 2017b Universal haplotype-based noninvasive prenatal testing for single gene diseases. Clinical Chemistry 63 513–524. (https://doi.org/10.1373/clinchem.2016.268375)
Jang SS, Lim BC, Yoo SK, Shin JY, Kim KJ, Seo JS, Kim JI & Chae JH 2018 Targeted linked-read sequencing for direct haplotype phasing of maternal DMD alleles: a practical and reliable method for noninvasive prenatal diagnosis. Scientific Reports 8 8678. (https://doi.org/10.1038/s41598-018-26941-0)
Jenkins LA, Deans ZC, Lewis C & Allen S 2018 Delivering an accredited non-invasive prenatal diagnosis service for monogenic disorders and recommendations for best practice. Prenatal Diagnosis 38 44–51. (https://doi.org/10.1002/pd.5197)
Kuhlmann-Capek M, Chiossi G, Singh P, Monsivais L, Lozovyy V, Gallagher L, Kirsch N, Florence E, Petruzzi V & Chang J et al. 2019 Effects of medication intake in early pregnancy on the fetal fraction of cell-free DNA testing. Prenatal Diagnosis 39 361–368. (https://doi.org/10.1002/pd.5436)
Lau TK, Cheung SW, Lo PS, Pursley AN, Chan MK, Jiang F, Zhang H, Wang W, Jong LF & Yuen OK et al. 2014 Non-invasive prenatal testing for fetal chromosomal abnormalities by low-coverage whole-genome sequencing of maternal plasma DNA: review of 1982 consecutive cases in a single center. Ultrasound in Obstetrics & Gynecology 43 254-264. (https://doi.org/10.1002/uog.13277)
Lenaerts L, Van Calsteren K, Che H, Vermeesch JR & Amant F 2019 Pregnant women with confirmed neoplasms should not have noninvasive prenatal testing. Prenatal Diagnosis 39 1162–1165. (https://doi.org/10.1002/pd.5544)
Lewis C, Hill M & Chitty LS 2017 Offering non-invasive prenatal testing as part of routine clinical service. Can high levels of informed choice be maintained? Prenatal Diagnosis 37 1130–1137. (https://doi.org/10.1002/pd.5154)
Lo YM, Corbetta N, Chamberlain PF, Rai V, Sargent IL, Redman CW & Wainscoat JS 1997 Presence of fetal DNA in maternal plasma and serum. Lancet 350 485–487. (https://doi.org/10.1016/S0140-6736(9702174-0)
Lo YM, Zhang J, Leung TN, Lau TK, Chang AM & Hjelm NM 1999 Rapid clearance of fetal DNA from maternal plasma. American Journal of Human Genetics 64 218–224. (https://doi.org/10.1086/302205)
Lo YM, Tsui NB, Chiu RW, Lau TK, Leung TN, Heung MM, Gerovassili A, Jin Y, Nicolaides KH & Cantor CR et al. 2007 Plasma placental RNA allelic ratio permits noninvasive prenatal chromosomal aneuploidy detection. Nature Medicine 13 218–223. (https://doi.org/10.1038/nm1530)
Lo YM, Chan KC, Sun H, Chen EZ, Jiang P, Lun FM, Zheng YW, Leung TY, Lau TK & Cantor CR et al. 2010 Maternal plasma DNA sequencing reveals the genome-wide genetic and mutational profile of the fetus. Science Translational Medicine 2 61ra91. (https://doi.org/10.1126/scitranslmed.3001720)
Lun FM, Tsui NB, Chan KC, Leung TY, Lau TK, Charoenkwan P, Chow KC, Lo WY, Wanapirak C & Sanguansermsri T et al. 2008 Noninvasive prenatal diagnosis of monogenic diseases by digital size selection and relative mutation dosage on DNA in maternal plasma. PNAS 105 19920–19925. (https://doi.org/10.1073/pnas.0810373105)
Minear MA, Lewis C, Pradhan S & Chandrasekharan S 2015 Global perspectives on clinical adoption of NIPT. Prenatal Diagnosis 35 959–967. (https://doi.org/10.1002/pd.4637)
Moise KJ, Jr, Hashmi SS, Markham K, Argoti PS & Bebbington M 2019 Cell free fetal DNA to triage antenatal rhesus immune globulin: is it really cost-effective in the United States? Prenatal Diagnosis 39 238–247. (https://doi.org/10.1002/pd.5415)
New MI, Tong YK, Yuen T, Jiang P, Pina C, Chan KC, Khattab A, Liao GJ, Yau M & Kim SM et al. 2014 Noninvasive prenatal diagnosis of congenital adrenal hyperplasia using cell-free fetal DNA in maternal plasma. Journal of Clinical Endocrinology and Metabolism 99 E1022–E1030. (https://doi.org/10.1210/jc.2014-1118)
Norton ME, Jacobsson B, Swamy GK, Laurent LC, Ranzini AC, Brar H, Tomlinson MW, Pereira L, Spitz JL & Hollemon D et al. 2015 Cell-free DNA analysis for noninvasive examination of trisomy. New England Journal of Medicine 372 1589–1597. (https://doi.org/10.1056/NEJMoa1407349)
Pan M, Li FT, Li Y, Jiang FM, Li DZ, Lau TK & Liao C 2013 Discordant results between fetal karyotyping and non-invasive prenatal testing by maternal plasma sequencing in a case of uniparental disomy 21 due to trisomic rescue. Prenatal Diagnosis 33 598–601. (https://doi.org/10.1002/pd.4069)
Parks M, Court S, Cleary S, Clokie S, Hewitt J, Williams D, Cole T, Macdonald F, Griffiths M & Allen S 2016 Non-invasive prenatal diagnosis of Duchenne and Becker muscular dystrophies by relative haplotype dosage. Prenatal Diagnosis 36 312–320. (https://doi.org/10.1002/pd.4781)
Parks M, Court S, Bowns B, Cleary S, Clokie S, Hewitt J, Williams D, Cole T, Macdonald F & Griffiths M et al. 2017 Non-invasive prenatal diagnosis of spinal muscular atrophy by relative haplotype dosage. European Journal of Human Genetics 25 416–422. (https://doi.org/10.1038/ejhg.2016.195)
Pertile MD, Halks-Miller M, Flowers N, Barbacioru C, Kinnings SL, Vavrek D, Seltzer WK & Bianchi DW 2017 Rare autosomal trisomies, revealed by maternal plasma DNA sequencing, suggest increased risk of feto-placental disease. Science Translational Medicine 9 eaan1240. (https://doi.org/10.1126/scitranslmed.aan1240)
Pescia G, Guex N, Iseli C, Brennan L, Osteras M, Xenarios I, Farinelli L & Conrad B 2017 Cell-free DNA testing of an extended range of chromosomal anomalies: clinical experience with 6,388 consecutive cases. Genetics in Medicine 19 169-175. (https://doi.org/10.1038/gim.2016.72)
Petersen AK, Cheung SW, Smith JL, Bi W, Ward PA, Peacock S, Braxton A, Van Den Veyver IB & Breman AM 2017 Positive predictive value estimates for cell-free noninvasive prenatal screening from data of a large referral genetic diagnostic laboratory. American Journal of Obstetrics and Gynecology 217 691.e1–691.e6. (https://doi.org/10.1016/j.ajog.2017.10.005)
Reiss RE, Discenza M, Foster J, Dobson L & Wilkins-Haug L 2017 Sex chromosome aneuploidy detection by noninvasive prenatal testing: helpful or hazardous? Prenatal Diagnosis 37 515–520. (https://doi.org/10.1002/pd.5039)
Scott F, Bonifacio M, Sandow R, Ellis K, Smet ME & Mclennan A 2018 Rare autosomal trisomies: important and not so rare. Prenatal Diagnosis 38 765–771. (https://doi.org/10.1002/pd.5325)
Sun K, Jiang P, Wong AIC, Cheng YKY, Cheng SH, Zhang H, Chan KCA, Leung TY, Chiu RWK & Lo YMD 2018 Size-tagged preferred ends in maternal plasma DNA shed light on the production mechanism and show utility in noninvasive prenatal testing. PNAS 115 E5106–E5114. (https://doi.org/10.1073/pnas.1804134115)
Tsao DS, Silas S, Landry BP, Itzep NP, Nguyen AB, Greenberg S, Kanne CK, Sheehan VA, Sharma R & Shukla R et al. 2019 A novel high-throughput molecular counting method with single base-pair resolution enables accurate single-gene NIPT. Scientific Reports 9 14382. (https://doi.org/10.1038/s41598-019-50378-8)
Tsui NB, Chiu RW, Ding C, El-Sheikhah A, Leung TN, Lau TK, Nicolaides KH & Lo YM 2005 Detection of trisomy 21 by quantitative mass spectrometric analysis of single-nucleotide polymorphisms. Clinical Chemistry 51 2358–2362. (https://doi.org/10.1373/clinchem.2005.056978)
Tsui NB, Kadir RA, Chan KC, Chi C, Mellars G, Tuddenham EG, Leung TY, Lau TK, Chiu RW & Lo YM 2011 Noninvasive prenatal diagnosis of hemophilia by microfluidics digital PCR analysis of maternal plasma DNA. Blood 117 3684–3691. (https://doi.org/10.1182/blood-2010-10-310789)
Van Der Meij KRM, Sistermans EA, Macville MVE, Stevens SJC, Bax CJ, Bekker MN, Bilardo CM, Boon EMJ, Boter M & Diderich KEM et al. 2019 TRIDENT-2: national implementation of genome-wide non-invasive prenatal testing as a first-tier screening test in the Netherlands. American Journal of Human Genetics 105 1091–1101. (https://doi.org/10.1016/j.ajhg.2019.10.005)
Van Opstal D, Van Maarle MC, Lichtenbelt K, Weiss MM, Schuring-Blom H, Bhola SL, Hoffer MJV, Huijsdens-Van Amsterdam K, Macville MV & Kooper AJA et al. 2018 Origin and clinical relevance of chromosomal aberrations other than the common trisomies detected by genome-wide NIPS: results of the TRIDENT study. Genetics in Medicine 20 480-485. (https://doi.org/10.1038/gim.2017.132)
Verhoef TI, Hill M, Drury S, Mason S, Jenkins L, Morris S & Chitty LS 2016 Non-invasive prenatal diagnosis (NIPD) for single gene disorders: cost analysis of NIPD and invasive testing pathways. Prenatal Diagnosis 36 636–642. (https://doi.org/10.1002/pd.4832)
Vermeulen C, Geeven G, De Wit E, Verstegen MJAM, Jansen RPM, Van Kranenburg M, De Bruijn E, Pulit SL, Kruisselbrink E & Shahsavari Z et al. 2017 Sensitive monogenic noninvasive prenatal diagnosis by targeted haplotyping. American Journal of Human Genetics 101 326–339. (https://doi.org/10.1016/j.ajhg.2017.07.012)
Verweij EJ, De Boer MA & Oepkes D 2014 Non-invasive prenatal testing for trisomy 13: more harm than good? Ultrasound in Obstetrics and Gynecology 44 112–114. (https://doi.org/10.1002/uog.13388)
Vossaert L, Wang Q, Salman R, Zhuo X, Qu C, Henke D, Seubert R, Chow J, U'ren L & Enright B et al. 2018 Reliable detection of subchromosomal deletions and duplications using cell-based noninvasive prenatal testing. Prenatal Diagnosis 38 1069–1078. (https://doi.org/10.1002/pd.5377)
Wan J, Li R, Zhang Y, Jing X, Yu Q, Li F, Li Y, Zhang L, Yi C & Li J et al. 2018 Pregnancy outcome of autosomal aneuploidies other than common trisomies detected by noninvasive prenatal testing in routine clinical practice. Prenatal Diagnosis 38 849-857. (https://doi.org/10.1002/pd.5340)
Wilkie AOM & Goriely A 2017 Gonadal mosaicism and non-invasive prenatal diagnosis for ‘reassurance’ in sporadic paternal age effect (PAE) disorders. Prenatal Diagnosis 37 946–948. (https://doi.org/10.1002/pd.5108)
Xiong L, Barrett AN, Hua R, Tan TZ, Ho SS, Chan JK, Zhong M & Choolani M 2015 Non-invasive prenatal diagnostic testing for beta-thalassaemia using cell-free fetal DNA and next generation sequencing. Prenatal Diagnosis 35 258–265. (https://doi.org/10.1002/pd.4536)
Xiong L, Barrett AN, Hua R, Ho S, Jun L, Chan K, Mei Z & Choolani M 2018 Non-invasive prenatal testing for fetal inheritance of maternal beta-thalassaemia mutations using targeted sequencing and relative mutation dosage: a feasibility study. BJOG 125 461–468. (https://doi.org/10.1111/1471-0528.15045)
Zhang B, Zhou Q, Chen Y, Shi Y, Zheng F, Liu J & Yu B 2019a High false-positive non-invasive prenatal screening results for sex chromosome abnormalities: are maternal factors the culprit? Prenatal Diagnosis In press. (https://doi.org/10.1002/pd.5529)
Zhang J, Li J, Saucier JB, Feng Y, Jiang Y, Sinson J, Mccombs AK, Schmitt ES, Peacock S & Chen S et al. 2019b Non-invasive prenatal sequencing for multiple Mendelian monogenic disorders using circulating cell-free fetal DNA. Nature Medicine 25 439–447. (https://doi.org/10.1038/s41591-018-0334-x)