The Next Generation of Prenatal Diagnosis: Genome-Wide DNA Sequencing in the Prenatal Period

Learning Objectives: After participating in this continuing professional development activity, the provider should be better able to: Review current prenatal diagnosis paradigms. Differentiate between exome and genome sequencings. Discuss limitations of genome-wide DNA sequencing. Prenatal genetic testing is advancing rapidly. A variety of prenatal diagnostic tests are available to identify different genetic and genomic abnormalities, and selecting the appropriate test requires careful collaboration. Next-generation sequencing (NGS) technologies—including exome sequencing (ES) and genome sequencing (GS)—have enabled the detection of single-gene disorders that are beyond the reach of traditional testing modalities such as karyotype and chromosomal microarray. NGS approaches can analyze varying amounts of genomic information, ranging from single genes to multigene panels, or focusing on the protein-coding regions of the genome. Despite its diagnostic power, ES and GS present several clinical and ethical challenges, including long turnaround time and high cost. Comprehensive pre- and posttest counseling is critical. This review focuses on the use of advanced sequencing technologies in anomalous fetuses. Introduction Prenatal genetic testing has become an integral part of clinical care for counseling pregnant individuals.1 In the United States, birth defects affect approximately 3% to 4% of births and are responsible for nearly 20% of infant mortality.2 A variety of prenatal diagnostic tests are available to identify different genetic and genomic abnormalities, and selecting the appropriate test requires careful collaboration among patients, genetics-trained providers, and genetic counselors. Traditionally, chromosomal abnormalities were identified using karyotyping; however, this method has been largely supplanted by chromosomal microarray analysis (CMA), which offers higher resolution for detecting submicroscopic deletions and duplications. More recently, NGS technologies—including ES and GS—have enabled the detection of single-gene disorders that are beyond the reach of CMA (Figure 1). NGS approaches can analyze varying amounts of genomic information, ranging from single genes to multigene panels or focusing on the protein-coding regions of the genome.3 Although whole genome sequencing (WGS) offers the most comprehensive view by examining the entire genome, ES and targeted gene panels are more commonly used in clinical settings due to lower cost and reduced DNA requirements.Figure 1.: Comparison of Prenatal Genetic Testing Options.Despite its diagnostic power, ES and GS present several clinical and ethical challenges. Turnaround time for results can take 3 to 4 weeks, and the high cost may limit access and exacerbate existing health care disparities. Additionally, both ES and GS frequently identify variants of uncertain significance (VUS) or incidental findings, which complicate result interpretation and counseling and may contribute to patient anxiety.4 Although few studies have reported on diagnostic yield in structurally normal fetuses5-7 (Table 1), this review will focus on the use of advanced sequencing technologies in anomalous fetuses, as current guidelines do not support the use of ES/GS in structurally normal fetuses at this time.8 Table 1. - Selected Studies Reporting on Diagnostic Yield of Exome and Genome Sequencing in Prenatal Diagnosis Reference Number of cases Inclusion criteria Method Positive/definite diagnoses Comments Exome sequencing Normand et al, 2018 146 Any anomaly Trio ES (39%) 32% Higher yield with multisystem abnormalities and craniofacial abnormalities Petrovski et al, 2019 234 Prospective; any anomaly Trio ES 10% Higher yield with multisystem organ abnormalities Lord et al, 2019 610 Prospective; any anomaly Trio ES 8.5% Higher yield with multisystem organ abnormalities Vora et al, 2020 102 Prospective and retrospective Trio ES 20.6% Higher yield given majority of cohort had multisystem anomalies Sparks et al, 2020 127 Nonimmune hydrops Trio ES (90%) 29% 30% of diagnoses were RASopathies Lei et al, 2020 163 Congenital anomalies of the kidney and urinary tract with or without other abnormalities Trio ES 12.3% Higher yield in multisystem abnormalities (27%) Li et al, 2020 260 Congenital heart defects Trio ES 10% Higher yield in cardiac rhabdomyoma, complex CHDs, septal defects Yaron et al, 2022 86 Central nervous abnormalities Trio ES 44% Highest diagnostic yield was noted in cases categorized as multisystem and complex brain anomalies Mau-Them et al, 2023 150 Prospective; ≥2 US anomalies or one US anomaly with high suspicion for genetic disorder Trio ES 41% Cases selected with high pretest probability for underlying genetic condition Jin et al, 2024 512 Retrospective, structural anomalies Trio ES 13.5% Highest diagnostic yield in fetuses with multisystem anomalies and skeletal differences Genome sequencing Choy et al, 2019 50 Increased NT +/- structural anomaly Proband GS 32% Twofold increase diagnostic yield compared with routine CMA/karyotype Zou et al, 2022 111 Prospective, any anomaly Trio GS 19.8% GS required less DNA and faster turnaround time compared with CMA + ES Hu et al, 2023 185 Prospective, any anomaly Trio GS (83%) 16.8% Increased diagnostic yield of 5.9% over CMA alone Abbreviations: CMA, chromosomal microarray analysis; ES, exome sequencing; GS, genome sequencing. Traditional Testing Paradigms Historically, prenatal diagnosis focused on the identification of fetal aneuploidy due to its high incidence and ease in diagnosis. Approximately 5% of clinically recognized pregnancies are aneuploid. The frequency is known to decrease as pregnancy progresses because of increased risk for spontaneous loss. Increasing maternal age is a well-established risk factor for aneuploidy.9 Karyotype was the gold standard for prenatal diagnosis for several decades. In 2012, Wapner et al reported on the increased diagnostic yield of chromosomal microarray (CMA).10 CMA allows for the detection of smaller copy number variants (CNVs) that would be missed by traditional karyotype analysis. Studies have demonstrated that the prevalence of early-onset disorders caused by submicroscopic CNVs is approximately 1 in 270.11 Unlike chromosomal aneuploidy, the risk for CNVs does not increase with maternal age. CMA is now considered the first-line testing for fetuses with structural anomalies identified on ultrasound. However, when CMA was nondiagnostic, fetuses with anomalies did not receive a genetic diagnosis before the use of sequencing technologies. Molecular Testing Techniques It is estimated that the human genome contains approximately 20,000 protein-coding genes.12 Of these genes, less than half are known to cause disease in humans.13 Traditionally, testing for single-gene disorders in pregnancy was only possible with a known family history of a monogenic disorder. However, as NGS technologies have advanced, our ability to diagnose single-gene disorders in fetuses has improved. Testing techniques include targeted gene sequencing for phenotypes associated with a specific panel of genes and genome-wide DNA sequencing to broadly examine either the exome or genome (Table 2). Table 2. - Molecular Testing Techniques: A Comparison Feature Gene sequencing Exome sequencing Genome sequencing Definition Sequences known genes Sequences the protein-coding regions of genes (exons) Sequences the entire genome, including exons, introns, and noncoding regions Sequencing depth >500× ~50 to 100× ~30 to 50× Cost Lowest (~$200 to $1000) Lower (~$400 to $1,500) Higher (~$1,000 to $3,000) Data size Smallest, easiest to analyze Smaller, easier to analyze Large, requiring more storage and analysis time Diagnostic utility Effective for detecting variants in known genes. Can identify SNPs, INDELs Effective for detecting SNVs in coding regions linked to genetic diseases Detects SNVs in coding and noncoding regions, structural variations, insertions/deletions, and variants in regulatory regions Turnaround time ~4 to 8 weeks ~4 to 8 weeks ~4 to 8 weeks Best for Identifying genetic disorders with known phenotype Identifying known genetic disorders affecting proteins Comprehensive genetic analysis, including rare variants and regulatory mutations Limitations Only able to detect changes in known genes Unable to detect variants in noncoding regions More complex data interpretation and storage needs Abbreviations: INDELs, insertions and deletions; SNVs, single nucleotide variants. Targeted gene sequencing Targeted gene panels were developed to test specific genes or sets of genes that are known to be associated with clinical phenotypes or known genetic syndromes such as skeletal dysplasia, cystic fibrosis, or nonimmune fetal hydrops. The benefit of targeted gene panels includes the ability to provide more comprehensive analysis of selected genes with high coverage and a lower likelihood of identifying VUS as compared with genome-wide DNA sequencing. Targeted gene panels are a reasonable starting point from a cost perspective with well-defined phenotypes. However, in one study examining the cost-effectiveness of targeted panels compared with ES for patients with nonimmune hydrops, ES was found to be the most cost-effective strategy.14 It is important to note that the genotype-phenotype correlation for genetic conditions identified in the prenatal period remains limited.15 For prenatal diagnosis, the fetal phenotype is determined from imaging studies, including ultrasound and MRI. There are many factors that can limit the quality of prenatal imaging, including early gestational age, fetal position, and maternal body acoustics. As such, determining the best targeted gene sequencing to order can be challenging. Because prenatal phenotypes are so limited, the approach to genetic testing is beginning to shift away from the phenotype-first strategy of gene panels toward more broad testing using genome-wide DNA sequencing. Targeted gene sequencing and gene panels are likely to become less useful as sequencing technology continues to improve. Exome sequencing ES describes the process of analyzing the protein-coding regions of the genome, referred to as the exome. The exome comprises approximately 1% to 2% of the genome and contains about 85% of known disease-causing genetic variants.16 In the postnatal setting, ES is a frequent tool for genetic diagnosis of children with congenital anomalies and neurodevelopmental delay.17,18 In prenatal diagnosis, ES can be considered for fetuses with anomalies suggestive of a genetic syndrome with a nondiagnostic CMA. The diagnostic yield for ES is higher in fetuses with multiple anomalies and in preselected populations with high pretest probability for having a monogenetic condition.19-28 A recent systematic review and meta-analysis, which included 72 reports, reported a diagnostic yield of ES for fetal structural anomalies ranging between 5% (95% CI -8% to 18%) and 89% (95% CI 72%–99%). The pooled incremental yield of ES over CMA/karyotype from all studies was 31% (95% CI 26%–36%). Notably, between study heterogeneity was high, and possible publication bias was reported.29 Trio-based (fetus and both biological parents) sequencing is recommended over singleton (fetus only) or duo (fetus and biological mother) sequencing. Trio analysis is associated with increased diagnostic yield.30 This testing strategy also improves interpretation of results, allowing identification of de novo variants, determining inheritance of recessive variants, and filtering of uninformative variants. Although trios are optimal, analysis pipelines for duos must be improved to allow families where the biological father is not available to provide a sample, the option of accessing ES. ES has been shown to have increased diagnostic yield than CMA alone, but it cannot identify all variants. ES is unable to detect variants in the noncoding (intronic) regions of the genome. This limits this technology's ability to detect structural variants (deletions or duplications), triple repeat disorders, variants in regulatory regions of the genome, and CNVs. Although the analysis of CNVs is improving, CNVs cannot to reliably detected through ES due to the nonuniform coverage of the exomes and variable read depth. At this time, a CMA should ideally be completed in conjunction with ES. Genome sequencing GS allows for an even more broad approach for genetic testing. GS describes sequencing the whole genome, including noncoding introns. GS analyzes most of the 3 billion base pairs of the genome. GS can detect aneuploidies, CNVs, single nucleotide variants (SNVs), insertions and deletions (INDELs), trinucleotide repeat expansions, intronic variants, and mitochondrial genome variants in a single testing platform. Fewer studies have been published on the detection rates for prenatal GS. One study reported a diagnostic yield of 15.1% in fetuses with structural anomalies, which was an incremental yield of 5.9% (11/185) over CMA alone.31 Prior studies reported similar diagnostic yields of 19.8%32 and 18.9%.33 Unpublished data from a multicenter study showed additional diagnostic yield of 12.6% in single anomalies and 26.2% in multiple anomalies when CMA was nondiagnostic.34 Although GS may have higher diagnostic rates, it also has higher incidence of identifying VUS and possibility incidental and secondary findings. GS also creates a large volume of data with less developed bioinformatic tools that limit its clinical utility currently. However, it is predicted that eventually GS will replace ES and CMA in prenatal diagnosis. Variant Classification Variations within individual human genome are common. When compared with the human genome reference, an exome sequence typically contains around 40,000 variants, while a genome sequence may have close to 3 million.35 The challenge lies in classifying these variants. The standard system for classifying genetic variants was developed by the American College of Medical Genetics and Genomics (ACMG) in collaboration with the Association for Molecular Pathology (AMP).36,37 These recommendations provide a standardized framework for clinicians to analyze the numerous variations found in sequencing and classify them as pathogenic or benign based on available evidence. The ACMG recommendations classify variants into five categories: pathogenic, likely pathogenic, VUS, likely benign, and benign. To determine classification, the ACMG has also established 28 criteria, incorporating various types of evidence, including population data, computational data, functional data, and segregation data. Variants are assessed based on these criteria, which are then used to assign a classification and determine the probability of pathogenicity.36 Variants are classified based on weight and type of evidence, with different levels of confidence for benign and pathogenic findings (Table 3). Table 3. - Classification system for genetic/genomic variants Classification Interpretation Pathogenic Well-established in current literature as disease-causing. Likely pathogenic Shows a genotype-phenotype correlation based on available data. Strong evidence suggests it is disease-causing, but not sufficient to confirm pathogenicity. Uncertain significance Cannot be definitively classified as benign or pathogenic due to limited data or variable phenotypic outcomes. Further evidence may lead to reclassification. Likely benign Strong evidence suggests the variant is not disease-causing, though insufficient data exist for a definitive benign classification. Benign Well-established in current literature as not disease-causing. Considered a negative result. Data from Richards S, Aziz N, Bale S, et al. Standards and guidelines for the interpretation of sequence variants: a joint consensus recommendation of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology. Genet Med Off J Am Coll Med Genet. 2015;17(5):405-424. doi:10.1038/gim.2015.30 Analysis of ES/GS results is phenotype-driven. For laboratories to correctly classify variants, complete information regarding imaging findings, pregnancy history, and medical family history is required. Additionally, identifying a pathogenic variant does not confirm a clinical diagnosis of the associated genetic condition. Interpretation of ES/GS results requires consideration of the phenotype in association with previously reported phenotypes. Variants of Uncertain Significance When a variant is identified that has an increased probability of pathogenicity, but findings do not clearly prove it as benign and without health consequences, the variant is classified as a VUS. Interpreting VUS can be even more complex in the prenatal setting. When an uncertain result is returned from genome-wide sequencing, pregnant individuals may require additional counseling and support. Results must be reviewed by a provider with significant genetic expertise. Prior studies have demonstrated that patients who understand that genome-wide sequencing results may be uncertain are more accepting of ambiguous results.38 Secondary Findings ES/GS can identify variants unrelated to the primary reason for testing; these are termed secondary, incidental, or medically actionable findings.3,39 The ACMG recommends evaluating 73 medically actionable genes during ES and GS.40 This approach aims to identify highly penetrant genetic disorders to reduce morbidity and mortality; however, patients may opt out of receiving these results.40? In the prenatal context, the disclosure of secondary findings presents unique ethical challenges. The ACMG currently does not provide specific recommendations for sequencing and reporting secondary findings in prenatal settings but suggests comprehensive discussions during pretest counseling.3,41 This counseling should also address the potential discovery of unexpected familial relationships, such as nonpaternity or consanguinity. The debate on reporting secondary findings in prenatal sequencing is ongoing. Some argue that reporting pathogenic variants in certain genes could inform prenatal care, while others express concerns about the ethical implications and the current lack of consensus. To navigate these challenges, patients should discuss their preferences and understanding of possible testing outcomes with a multidisciplinary team before undergoing genetic testing. Pretest and Posttest Counseling Prenatal genome-wide DNA sequencing is a powerful tool for clinicians, but it should be paired with comprehensive pre- and posttest counseling to ensure patients understand both the benefits and limitations of the technology. Counseling is critical and should address potential outcomes, including the possibility of identifying variants of uncertain significance (VUS), incidental findings, the option to decline testing for medically actionable genes as recommended by the American College of Medical Genetics (ACMG), result turnaround times, and options for data sharing and reanalysis.42 Additionally, patients must be counseled that a negative result does not exclude all genetic conditions. These conversations should involve a multidisciplinary team trained in discussing the medical, psychological, and economic implications of genetic testing during pregnancy. Shared decision-making is essential when considering more advanced sequencing techniques (Table S1, available online: https://links.lww.com/PGO/A14 for more detail).43,44 Clinical Utility As with any new technology, it is important to understand the impact that advanced sequencing has on clinical management. In prenatal diagnosis, genetic test results may inform a patient's decisions regarding pregnancy management. Decisions can include pregnancy termination, at a care care of the in and to or risk is in this including studies that of patients who to with prenatal genome-wide sequencing. Testing pregnant patients with an ultrasound anomaly should be genetic counseling and the option to diagnostic The to specific test should be based on the individual patient's family history, and specific findings on ultrasound. of prenatal genetic testing. At this time, a approach is genetic testing with a CMA with to to genome-wide DNA sequencing, results are However, with in GS, it is that GS may be the first-line genetic test in prenatal diagnosis. The ACMG suggests that ES can be considered for a with ultrasound anomalies and nondiagnostic CMA/karyotype that analysis be completed to diagnostic also the of pretest counseling with a provider who has in which must include information about secondary findings, incidental findings, and of results will be At the time of the GS was not recommended due to lack of clinical data and data However, new data on utility of prenatal GS is and guidelines may based on published data. The of Prenatal Diagnosis by the for has published guidelines to genome-wide DNA sequencing in the prenatal The are as for prenatal a current pregnancy with a with a single anomaly or with multiple organ system anomalies suggestive of a possible genetic but nondiagnostic or with CMA result multidisciplinary review in which the fetal phenotype suggests a single-gene a with an anomaly or anomalies a genetic and a of similar anomalies in the current pregnancy. guidelines also use of analysis. Additionally, do not the use of genome-wide DNA sequencing in fetuses without structural anomalies DNA sequencing is into the prenatal genetic testing. ES/GS will likely to as additional and data is including in accessing this technology, pre- and posttest variant interpretation with limited prenatal and current high cost of testing.