Noninvasive prenatal testing
Noninvasive prenatal testing (NIPT) is a screening technique that examines cell-free fetal DNA circulating in a pregnant woman's bloodstream to assess risks of common fetal chromosomal aneuploidies, including trisomy 21 (Down syndrome), trisomy 18 (Edwards syndrome), and trisomy 13 (Patau syndrome), without requiring invasive procedures like amniocentesis or chorionic villus sampling.[1][2] First commercialized in 2011 following advances in next-generation sequencing of cell-free fetal DNA discovered in maternal plasma, NIPT rapidly became a preferred first-line option over traditional biochemical and ultrasound-based screenings due to its superior performance, achieving detection rates over 99% for trisomy 21 with false-positive rates under 0.1% in high-risk pregnancies.[3][4][5] By reducing reliance on invasive diagnostics—which carry miscarriage risks of about 0.1-0.5%—NIPT has lowered overall procedure volumes while increasing screening uptake, though empirical data indicate variable impacts on termination rates for aneuploidy-affected pregnancies, with live birth prevalence of Down syndrome showing no consistent decline post-implementation in several jurisdictions.[6][7] Limitations persist, as NIPT's accuracy diminishes for sex chromosome aneuploidies, mosaicism, or maternal conditions mimicking fetal signals, yielding false positives that necessitate confirmatory testing and can provoke anxiety or erroneous decisions; reported false-negative rates may also be underestimated.[8][9] Ethical debates center on its potential to facilitate selective terminations, expand to microdeletions or non-medical traits like fetal sex with lower predictive values, and inadvertently pressure parents toward ending viable pregnancies, raising questions about equity, informed consent, and societal valuation of disabled lives amid commercial marketing that sometimes overstates diagnostic certainty.[10][11][12]History
Origins and Early Research
The pursuit of noninvasive prenatal diagnosis began in the mid-20th century with efforts to isolate rare fetal nucleated cells circulating in maternal blood, first reported in the 1960s.[13] These cells, estimated at frequencies of 1 in 10^6 to 10^9 maternal cells, were targeted for genetic analysis to avoid the risks of invasive procedures like amniocentesis, but isolation and enrichment proved technically challenging and yielded insufficient material for reliable testing.[13][14] A pivotal advancement occurred in 1997 when Y.M. Dennis Lo and colleagues at the University of Oxford demonstrated the presence of cell-free fetal DNA (cffDNA) in maternal plasma and serum.02174-0/fulltext) Using polymerase chain reaction (PCR) to amplify Y-chromosome-specific sequences (such as the SRY gene) from pregnancies with male fetuses, they detected fetal DNA in 46 of 60 tested samples (77%), with levels ranging from less than 1 to over 25 genome equivalents per milliliter of plasma.[15] This discovery established that cffDNA, derived primarily from placental trophoblast apoptosis, constitutes approximately 3-13% of total cell-free DNA in maternal circulation during the first and second trimesters, enabling potential noninvasive access to fetal genetic material without cell isolation.[13][16] Subsequent early research in the late 1990s and early 2000s focused on validating and applying cffDNA for specific diagnostics. Studies confirmed its detectability as early as 18 days post-conception and its utility for fetal sex determination (with >95% accuracy by quantitative real-time PCR) and noninvasive genotyping of single-nucleotide polymorphisms, such as for RhD status in RhD-negative mothers to guide anti-D prophylaxis.[15][14] By 2008, Lo's group advanced to aneuploidy detection, employing massively parallel shotgun sequencing of maternal plasma DNA to identify trisomy 21 in eight affected pregnancies with 100% sensitivity and specificity in proof-of-concept trials, measuring chromosomal representation by z-scores derived from read depth ratios.[17] Parallel efforts using digital PCR for relative quantification of chromosome 21 sequences further corroborated feasibility, though limited by the low fetal fraction and need for high-depth sequencing to distinguish euploid from aneuploid profiles amid maternal DNA background.[18] These foundational studies shifted noninvasive prenatal testing from qualitative genotyping to quantitative chromosomal analysis, setting the stage for broader clinical validation.[19]Commercial Introduction and Widespread Adoption
The first commercial noninvasive prenatal testing (NIPT) services emerged in 2011, with BGI launching the NIFTY test in Hong Kong in August and Sequenom introducing MaterniT21 in the United States in October, both utilizing massively parallel sequencing of cell-free fetal DNA to screen for common trisomies.[11][20] These launches followed foundational research demonstrating the feasibility of analyzing fetal DNA fractions in maternal plasma, initially targeting high-risk pregnancies defined by maternal age over 35 or abnormal serum screening results.[21] Early adoption was driven by the tests' reported sensitivity exceeding 99% for trisomy 21 and substantially lower false-positive rates compared to traditional serum-based aneuploidy screening methods like the quadruple test.[22] Subsequent commercialization accelerated with Verinata Health's Harmony test in 2012 (later acquired by Illumina) and Natera's Panorama in 2013, expanding the market through variations in analytical platforms such as targeted amplification and whole-genome sequencing.[23][20] Legal disputes, including Sequenom's patent infringement suits against competitors, temporarily shaped market dynamics but did not impede overall growth, as courts invalidated broad cfDNA sequencing claims by 2015.[23] By 2012, the global NIPT market was valued at $0.22 billion, reflecting initial penetration primarily in private sectors of North America, Europe, and Asia.[11] Widespread adoption gained momentum through endorsements from professional societies, initially restricting use to high-risk cases; for instance, the American College of Medical Genetics and Genomics (ACMG) supported NIPT as a primary screening option for high-risk women in 2012, followed by expansions.[24] The American College of Obstetricians and Gynecologists (ACOG) and Society for Maternal-Fetal Medicine (SMFM) updated guidelines in 2015 to endorse NIPT for average-risk pregnancies, and by 2020 recommended it for all pregnant individuals regardless of risk, citing improved detection rates and reduced need for invasive diagnostics.[25] In Europe, the European Society of Human Genetics (ESHG) and American Society of Human Genetics (ASHG) jointly recommended NIPT as a contingent test in 2015, though implementation varied by country due to regulatory and reimbursement differences.[26] Uptake statistics illustrate rapid market penetration: in the US, NIPT accounted for a growing share of prenatal screenings, with the market reaching $1.60 billion by 2022 amid approximately 3.7 million annual births, implying testing in a substantial fraction of pregnancies following guideline expansions.[27][28] Globally, services expanded to nearly 90 countries by the mid-2010s, with over two million tests performed cumulatively by the early 2020s, though utilization remained below 25% in most European nations as of 2020, higher in Italy and Spain due to private sector access.[29][30][31] This adoption was propelled by empirical validation studies confirming >99% positive predictive values for trisomy 21 in validation cohorts, alongside cost reductions from initial $2,000+ per test to under $1,000 by the late 2010s, though equity concerns persist in public health systems with limited reimbursement.[22][29]Recent Expansions and Technological Advances
In the early 2020s, noninvasive prenatal testing (NIPT) expanded beyond screening for common trisomies (13, 18, and 21) to include sex chromosome aneuploidies (SCAs) and rare autosomal trisomies (RATs), with commercial panels incorporating these by 2020-2021 to address gaps in traditional serum screening.[32] Further extensions targeted copy number variations (CNVs), such as microdeletions and microduplications, enabling detection of subchromosomal abnormalities like 22q11.2 deletion syndrome, though with variable positive predictive values requiring confirmatory diagnostics.[33] These expansions were driven by algorithmic refinements and higher sequencing depths, which improved resolution for smaller CNVs while maintaining low false-positive rates for aneuploidies.[34] Technological advances emphasized enhanced bioinformatics and increased read depths in massively parallel sequencing, allowing NIPT-plus platforms to achieve sensitivities over 99% for expanded panels, including all autosomes and select CNVs, as validated in studies from 2023 onward.[35] Integration of artificial intelligence for variant calling and noise reduction has further boosted accuracy, particularly for low-fetal-fraction samples, reducing reliance on invasive procedures like amniocentesis by up to 50% in screened populations.[36] Emerging cell-based NIPT (cbNIPT) approaches, leveraging intact fetal cells from maternal blood, promise whole-genome analysis without fragmentation biases inherent in cell-free DNA methods, with proof-of-concept data emerging around 2022.[37] By 2023-2025, focus shifted toward single-gene disorder (SGD) screening, with trials like Natera's EXPAND study—launched in 2023 and enrolling over 1,600 participants by October 2025—evaluating noninvasive detection of de novo or paternally inherited dominant variants in genes like those for Noonan syndrome and achondroplasia.[38] Commercial expansions, such as Myriad Genetics' January 2025 platform for targeted SGDs using advanced capture sequencing, reported sensitivities near 100% and specificities above 99.9% for select panels, though limited to high-risk cases due to technical challenges in haplotype phasing.[39] These developments underscore NIPT's evolution from broad aneuploidy screening to precise, genome-wide applications, tempered by ongoing needs for validation in diverse populations and ethical considerations around incidental findings.[40]Scientific and Technical Foundations
Cell-Free Fetal DNA and Detection Mechanisms
Cell-free fetal DNA (cffDNA) consists of short, fragmented DNA sequences originating primarily from apoptotic placental trophoblasts that enter the maternal bloodstream during pregnancy.[41] These fragments, typically less than 200 base pairs in length, become detectable in maternal plasma as early as the fourth week of gestation and constitute a small proportion of the total cell-free DNA (cfDNA), which is predominantly maternal in origin.[41] The presence of cffDNA in maternal circulation was first demonstrated in 1997 by Y.M. Dennis Lo and colleagues, who detected Y-chromosome sequences in plasma from women pregnant with male fetuses, drawing parallels to cell-free tumor DNA detection.[13] The proportion of cffDNA relative to total cfDNA, known as the fetal fraction, typically ranges from 4% to 10% after 10 weeks of gestation, increasing with advancing gestational age and influenced by factors such as maternal body mass index and placental mass.[41] A fetal fraction below 4% often renders noninvasive prenatal testing unreliable, leading to test failure rates of up to 3-5% in clinical practice, particularly in obese pregnancies or early gestation.[41] Accurate quantification of the fetal fraction is essential prior to analysis, achieved through methods like counting fetal-specific polymorphisms or chromosome Y reads in male pregnancies.[24] Detection begins with a maternal venous blood draw of 10-20 mL, followed by centrifugation to isolate plasma, from which cfDNA is extracted using silica-based columns or magnetic beads to preserve fragment integrity.[41] The extracted cfDNA undergoes library preparation and sequencing, with bioinformatics pipelines distinguishing fetal contributions by leveraging differences in fragment size, methylation patterns, or sequence content.[24] Post-delivery, cffDNA clears rapidly from maternal plasma, with a half-life of approximately 16 minutes, confirming its pregnancy-specific origin.[13] Primary detection mechanisms rely on massively parallel sequencing (MPS) of cfDNA to infer fetal chromosomal copy number. In whole-genome shotgun sequencing, millions of short reads (tens of millions total) are generated and mapped to a reference genome, with chromosomal aneuploidies detected by calculating the proportional representation of reads per chromosome, adjusted for GC bias via normalized chromosome values (NCVs) or Z-scores; for trisomy 21, an excess of approximately 50% reads from chromosome 21 indicates the condition.[24] This count-based approach achieves sensitivities of 98.6-100% and specificities of 99.8-100% for common trisomies in validation studies.[24] Targeted sequencing variants amplify and sequence only regions from chromosomes 13, 18, and 21, reducing data volume and costs while maintaining high accuracy for specific aneuploidies.[24] SNP-based methods, which genotype single nucleotide polymorphisms to model paternal and maternal contributions, enable detection of uniparental disomy or mosaicism but require higher fetal fractions for precision.[24] These techniques avoid physical separation of fetal and maternal DNA, relying instead on statistical modeling to deconvolute the mixed cfDNA pool.[24]Methodological Variations and Analytical Approaches
Noninvasive prenatal testing (NIPT) primarily utilizes next-generation sequencing (NGS) of cell-free fetal DNA (cffDNA) from maternal plasma, with methodological variations centered on sequencing scope and depth to detect chromosomal aneuploidies such as trisomies 13, 18, and 21. The predominant approaches include whole-genome sequencing (WGS), typically executed via massively parallel shotgun sequencing (MPSS) at low coverage (e.g., 0.1× to 0.3×), and targeted sequencing, which focuses on select chromosomal regions. These methods quantify fetal DNA contribution relative to maternal background, requiring fetal fraction estimates typically above 4% for reliable results.[24][42] In WGS/MPSS, millions of short cfDNA fragments (50-100 base pairs) are sequenced genome-wide without prior enrichment, mapping reads to reference genomes to count chromosomal representation. Analytical pipelines normalize read depths for biases like GC content and mappability, then apply statistical models—such as z-scores comparing observed-to-expected ratios or normalized chromosome values (NCVs)—to flag deviations exceeding thresholds (e.g., z-score >3 for trisomy). This unbiased approach, validated in studies with sensitivities of 98.6-100% for trisomy 21, generates substantial data (2.3-6.3 million mapped reads per sample) but demands high computational resources and multiplexing limits (e.g., 192 samples per run).[24][42][43] Targeted sequencing variants, including chromosome-selective sequencing (CSS) or directed amplification of nucleotide sequences and regions (DANSR), enrich cfDNA from high-risk loci (e.g., 576 sites on chromosomes 18 and 21), achieving deeper coverage with fewer reads (~1 million) for cost efficiency and faster processing. Analysis integrates selective read counts with fetal fraction metrics, often via Bayesian classifiers or allele ratio assessments, yielding comparable sensitivities (97-100%) but higher failure rates (~2-4%) due to enrichment biases or low fetal fraction. SNP-based targeted methods genotype polymorphic sites to differentiate fetal alleles, enhancing detection of mosaicism (<30%) and sex chromosome aneuploidies, though they risk no-call rates from maternal homozygosity.[42][43] For expanded applications like copy number variants (CNVs) and microdeletions, analytical approaches shift to higher-depth NGS or specialized pipelines, such as hidden Markov models for segmental imbalances, though resolution remains limited (~7-10 Mb) without whole-genome higher coverage. Non-sequencing alternatives include microarray-based quantification of cfDNA intensities or digital PCR for locus-specific counting, offering rapidity but scalability constraints and reduced sensitivity for low-level mosaicism. Fetal fraction calculation varies—e.g., Y-chromosome proportion in MPSS for male fetuses or SNP heterozygosity ratios in targeted assays—underpinning overall test validity across methods.[43][44]Targeted Conditions
Screening for Common Trisomies
Noninvasive prenatal testing (NIPT) screens for trisomy 21 (Down syndrome), trisomy 18 (Edwards syndrome), and trisomy 13 (Patau syndrome) by sequencing cell-free fetal DNA (cffDNA) fragments in maternal blood, which originate predominantly from placental trophoblasts, to detect chromosomal overrepresentation.[45] The primary analytical methods include massively parallel shotgun sequencing (MPSS), which counts DNA reads aligned to each chromosome, or targeted approaches like single-nucleotide polymorphism (SNP)-based analysis, both calibrated against euploid reference ratios to flag deviations exceeding predefined thresholds (typically Z-scores >3 for aneuploidy).[46] Testing is reliable from 10 weeks gestation onward, requiring a fetal fraction of at least 4%, with failure rates under 3% in most cohorts.[47] Meta-analyses of clinical performance data from over 100,000 pregnancies demonstrate high sensitivity for trisomy 21, pooled at 99% (95% CI: 98-99%), with specificity approaching 100% (99.9%, 95% CI: 99.9-100%); for trisomy 18, sensitivity averages 98% (95% CI: 96-99%), and for trisomy 13, 91% (95% CI: 87-94%), alongside specificities exceeding 99.9% for both.[48] [49] Individual validation studies corroborate these figures; for instance, a 2019 cohort of 1,457 Korean pregnancies reported 100% sensitivity and 99.9% specificity for trisomy 21, 92.9% sensitivity and 100% specificity for trisomy 18, and 100% sensitivity with 99.9% specificity for trisomy 13, though smaller sample sizes for rarer trisomies limit precision.[50] Positive predictive values (PPVs) vary by maternal age and population prevalence: in unselected cohorts, PPV for trisomy 21 reaches 82-91%, but drops to 40-68% for trisomy 18 and 9-28% for trisomy 13 due to their lower incidences (approximately 1:5,000-1:16,000 live births versus 1:700 for trisomy 21).[51]| Trisomy | Sensitivity (%) | Specificity (%) | PPV Range (General Population, %) | Primary Source |
|---|---|---|---|---|
| 21 (Down) | 99-100 | 99.9-100 | 82-91 | Meta-analysis of 112,628 cases[48] |
| 18 (Edwards) | 92-98 | 99.9-100 | 40-68 | Validation in 1,457 pregnancies; meta-analysis[50] [48] |
| 13 (Patau) | 90-100 | 99.9-100 | 9-28 | Meta-analysis; cohort study[48] [50] |
Detection of Rare Aneuploidies, Microdeletions, and Single-Gene Disorders
Expanded noninvasive prenatal testing (NIPT) panels have been developed to screen for rare aneuploidies beyond the common trisomies 21, 18, and 13, including autosomal trisomies such as those of chromosomes 7, 8, 9, 15, 16, and 22, as well as rare sex chromosome aneuploidies like 47,XXX, 47,XXY, 47,XYY, and 45,X.[53] These expanded approaches typically rely on genome-wide sequencing of cell-free fetal DNA (cffDNA) to identify copy number variations (CNVs), achieving detection rates exceeding 90% for some rare trisomies in validation studies, though positive predictive values (PPVs) remain low due to the infrequency of these conditions, often below 5%.[54] For instance, in a cohort analysis, the PPV for rare fetal aneuploidies detected by NIPT was reported as 4.90%, highlighting the predominance of false positives that necessitate confirmatory invasive testing like amniocentesis.[54] Sensitivity for specific rare aneuploidies, such as 45,X, has been estimated at around 90%, but overall performance varies with fetal fraction and sequencing depth, with false positives arising from confined placental mosaicism or maternal CNVs.[53][55] Detection of microdeletions and microduplications, such as 22q11.2 deletion syndrome (DiGeorge syndrome), 1p36 deletion syndrome, and cri-du-chat (5p deletion), represents a further extension of NIPT using methods like whole-genome shallow sequencing or targeted CNV analysis.[56] Studies have demonstrated sensitivities approaching 100% for certain clinically significant microdeletions in expanded NIPT when compared to chromosomal microarray analysis (CMA), with specificities also high in controlled settings; however, real-world PPVs for these rare events (prevalence often <1 in 2,000) can drop below 50%, leading to substantial overcalling of variants that are either benign or confined to the placenta.[53][56] Factors influencing accuracy include the size of the deletion (typically requiring >500 kb for reliable detection), fetal DNA fraction (>4%), and algorithmic thresholds, with performance degrading for smaller or mosaic variants.[57] The International Society for Prenatal Diagnosis has cautioned that expanded NIPT for microdeletions lacks sufficient prospective validation for routine use, as rarity precludes large-scale cohort studies, potentially resulting in unnecessary anxiety and procedures.[58] Screening for single-gene disorders via NIPT is an emerging application, primarily targeting de novo mutations or paternally inherited dominant conditions using relative haplotype dosage (RHD) or single-nucleotide polymorphism (SNP)-based analysis of cffDNA.[59] Commercial tests, such as those evaluating up to 202 dominant single-gene disorders, have reported PPVs greater than 90% for specific conditions like achondroplasia or thanatophoric dysplasia in high-risk pregnancies, but applicability is limited to cases without maternal inheritance risk, as maternal variants complicate haplotype phasing.[60] For recessive disorders like cystic fibrosis, noninvasive prenatal diagnosis requires prior parental genotyping and relative mutation dosage, achieving near-100% accuracy from 9 weeks gestation in research settings, though clinical implementation remains investigational and not broadly recommended due to technical complexity and validation gaps.[61][62] Overall, single-gene NIPT does not supplant invasive diagnostics and is constrained by low fetal fraction thresholds, bioinformatic challenges in variant calling, and ethical concerns over expanding screening to rare monogenic traits without established clinical utility.[63]Clinical Implementation
Testing Procedure and Timing
Noninvasive prenatal testing (NIPT), also known as cell-free DNA screening, involves a simple maternal blood draw to obtain cell-free fetal DNA (cffDNA) circulating in the pregnant woman's bloodstream.[64][65] The procedure typically requires 8-10 mL of venous blood collected via standard venipuncture, similar to routine prenatal blood tests, with no special preparation needed beyond fasting if required by the specific laboratory protocol.[66] The sample is then shipped to a certified laboratory, where cffDNA is isolated, amplified, and analyzed using techniques such as massively parallel sequencing or targeted PCR to detect chromosomal aneuploidies by quantifying fetal DNA fractions relative to maternal DNA.[67] Results are generally available within 5-10 days, depending on the testing platform and laboratory turnaround time.[64] Testing is recommended starting at 10 weeks of gestation to ensure adequate fetal fraction—the proportion of cffDNA in the maternal sample, typically requiring at least 4% for reliable results—as earlier sampling yields insufficient fetal DNA, increasing the risk of inconclusive outcomes.[68][64] It can be performed up to term, though most screenings occur in the first or second trimester, with median gestational age around 16 weeks in clinical studies.[69] For optimal accuracy, guidelines from bodies like the American College of Obstetricians and Gynecologists (ACOG) advise integrating NIPT with first-trimester ultrasound for dating and anomaly assessment, but the blood draw itself poses no procedural risks to the fetus.[70][71] Variations in procedure may include expanded panels for microdeletions or single-gene disorders, which require higher-resolution sequencing but follow the same blood collection method.[66] In cases of low fetal fraction, often linked to maternal obesity or early gestation, redraws may be advised after 1-2 weeks to allow fetal DNA levels to rise.[72] Pre-test genetic counseling is standard to discuss limitations, such as the screening nature of NIPT rather than diagnostic confirmation.[73]Accuracy Metrics, False Positives/Negatives, and Validation Studies
Noninvasive prenatal testing (NIPT) demonstrates high sensitivity and specificity for detecting common fetal trisomies, particularly trisomy 21 (Down syndrome), with reported sensitivity rates exceeding 99% and specificity above 99% in multiple large-scale studies.[49] For trisomy 18 (Edwards syndrome) and trisomy 13 (Patau syndrome), sensitivity is typically 96-98%, with specificity remaining over 99%, though performance can vary by platform and population.[49] Positive predictive value (PPV) for these conditions is influenced by maternal age and prior risk, often ranging from 80-95% for trisomy 21 in average-risk pregnancies, but dropping for rarer aneuploidies due to lower prevalence.[74] These metrics position NIPT as superior to traditional serum screening, which has false-positive rates around 5% for trisomy 21, compared to under 0.1% for NIPT.[2] False positives in NIPT primarily arise from confined placental mosaicism (CPM), where chromosomal abnormalities are present in the placenta but not the fetus, leading to discordant results upon confirmatory invasive testing like amniocentesis.[75] CPM accounts for a significant portion of discrepancies, with studies estimating it contributes to 20-50% of false positives for trisomies 18 and 13, though less frequently for trisomy 21.[76] Other contributors include maternal copy number variants, vanishing twins, or technical artifacts in cell-free DNA analysis, but overall false-positive rates remain low at 0.05-0.2% for trisomy 21 across validation cohorts.[77] Maternal factors such as obesity or low fetal fraction can exacerbate inaccuracies, prompting no-call rates of 2-5% in some tests.[2] False negatives are rarer, with sensitivity failures often linked to low fetal DNA fraction or mosaicism not captured in maternal plasma, though reported rates are under 0.5% for trisomy 21 and potentially underestimated in smaller studies.[8] A 2022 analysis suggested that while false-positive rates align with manufacturer claims, false-negative rates may be higher than reported due to under-ascertainment in low-prevalence settings.[8] Validation studies, including prospective trials and meta-analyses, underpin NIPT's reliability. A 2022 multicenter study of over 20,000 pregnancies reported detection rates of 99.2% for trisomy 21 and false positives of 0.09%, validated against karyotyping.[49] A 2024 evaluation in 37,891 cases confirmed overall sensitivity of 96.55% and specificity of 99.89% for trisomies 21, 18, and 13 using Z-score methods.[74] A 2025 meta-analysis of NIPT versus invasive procedures affirmed superior diagnostic performance, with pooled sensitivity for trisomy 21 at 99.7% and specificity at 99.9%, though emphasizing the need for confirmatory diagnostics due to residual error risks.[78] These studies highlight NIPT's evolution from initial 2011-2012 proofs-of-concept to genome-wide platforms, with ongoing refinements addressing mosaicism via algorithms like mosaicism ratios.[79]| Condition | Sensitivity (%) | Specificity (%) | False-Positive Rate (%) | Key Validation Source |
|---|---|---|---|---|
| Trisomy 21 | 99.2-100 | >99 | 0.05-0.09 | Frontiers in Genetics (2022)[49] |
| Trisomy 18 | 96.3-98 | >99 | 0.1-0.13 | Frontiers in Genetics (2022)[49] |
| Trisomy 13 | 96-98 | >99 | 0.1-0.2 | BMJ Genetics (2025)[74] |