Molecular diagnostic yield of exome sequencing in a Chinese cohort of 512 fetuses with anomalies

Abstract

Background

Currently, whole exome sequencing has been performed as a helpful complement in the prenatal setting in case of fetal anomalies. However, data on its clinical utility remain limited in practice. Herein, we reported our data of fetal exome sequencing in a cohort of 512 trios to evaluate its diagnostic yield.

Methods

In this retrospective cohort study, the couples performing prenatal exome sequencing were enrolled. Fetal phenotype was classified according to ultrasound and magnetic resonance imaging findings. Genetic variants were analyzed based on a phenotype-driven followed by genotype-driven approach in all trios.

Results

A total of 97 diagnostic variants in 65 genes were identified in 69 fetuses, with an average detection rate of 13.48%. Skeletal and renal system were the most frequently affected organs referred for whole exome sequencing, with the highest diagnostic rates. Among them, short femur and kidney cyst were the most common phenotype. Fetal growth restriction was the most frequently observed phenotype with a low detection rate (4.3%). Exome sequencing had limited value in isolated increased nuchal translucency and chest anomalies.

Conclusions

This study provides our data on the detection rate of whole exome sequencing in fetal anomalies in a large cohort. It contributes to the expanding of phenotypic and genotypic spectrum.

Introduction

Fetal anomalies occur in approximately 4–6% pregnancies [1], which result in increased perinatal morbidity and mortality, causing economic and mental burden on the families and society [2]. Ultrasound scanning is a preferred method to identify fetal anomalies, ranging from minor deformity to multiple anomalies [3, 4]. Some anomalies, such as ventricular septal defect and lymphatic hygroma, may have good prognoses, whereas, pathogenic genetic causes can yield contrary outcomes [5,6,7]. Therefore, genetic investigations are necessary in the assessment of fetal prognoses and clinical decision-making.

Common genetic causes including chromosomal abnormalities, copy number variations (CNVs) and single-nucleotide variants (SNVs) [8]. Generally, diagnostic yield for karyotyping is nearly 32% in fetuses with anomalies [9, 10]. Together with chromosomal microarray (CMA), detection rates can be increased by 3–10% [1, 3, 11]. Whole exome sequencing (WES) are recommended by the American College of Medical Genetics and Genomics (ACMG) (2020) as a helpful complement after negative findings in karyotyping and CMA [12]. Currently, publications about the application of fetal WES in prenatal diagnosis are increased, with a reported diagnostic yield varying from 6.1-44% [1, 3, 13,14,15,16,17,18,19]. Regretfully, only four cohorts are reported with relatively large sample size (n > 500) [1, 3, 13, 14]. Therefore, data on its clinical utility remain limited in practice.

Herein, we reported our data of fetal WES in a Chinese cohort of 512 trios to evaluate its detection value for fetal anomalies. Fetal phenotype was classified according to ultrasound and magnetic resonance imaging (MRI) findings and variants were analyzed with a phenotype-driven followed by genotype-driven strategy. This study will contribute to a more comprehensive understanding of the clinical efficiency of prenatal WES and expand the phenotypic and genotypic spectrum.

Methods

Study design and participants

It is a retrospective cohort study in a tertiary referral centre (Women’s Hospital, School of Medicine, Zhejiang University, Hangzhou, China), from November 1, 2021 to January 16, 2024. Clinical data were collected until April 23, 2024.

Invasive prenatal tests were performed in pregnant women whose fetuses manifesting as structural anomalies. The data interpretation was made during pregnancy, after a turnaround time of WES for nearly 2 weeks, and the couple might choose to continue or terminate the pregnancy, based on the WES results. In the current investigation, families in accordance with the following criteria were recruitted: (1) fetal anomalies were diagnosed by a qualified physician with ultrasound scanning or MRI findings; (2) fetal WES was performed and peripheral blood for both parents were available for further WES analysis. Exclusion criteria were pregnancies with aneuploidies or pathogenic copy number variations. Genetic and phenotypic data, as well as the outcome of the pregnancy of the couple were collected and analyzed.

Written informed consents were obtained and the study was approved by the Ethics Committee of Women’s Hospital, School of Medicine Zhejiang University (IRB-20240128-R).

DNA extraction

Fetal DNA was extracted from chorionic villi in three (0.59%) cases, amniotic fluid in 341 (66.6%), umbilical cord blood in 98 (19.14%) and tissue samples after miscarriage or termination of pregnancy (TOP) in 70 (13.67%). The average gestational age of villocentesis was 13.55 ± 0.51 weeks, amniocentesis, 22.54 ± 0.55 and transabdominal umbilical blood puncture, 29.55 ± 0.71 weeks. DNA was extracted using the QIAamp DNA Blood Mini Kit (Qiagen, Germany) according to the manufacturer’s protocol. The DNA concentrations were detected using the NanoDrop 2000 (Thermo Fisher Scientific, USA).

Whole exome sequencing

DNA samples from parents-fetus trios were sequenced on Illumina NovaSeq 6000 platform (Illumina, San Diego, CA, USA), according to the manufacturer’s instructions. The sequencing reads were referred to the Genome Reference Consortium Human genome build 37 (GRCh37) or humangenome 19 (hg19). Variants were filtered by population databases, including genome Aggregation Database (gnomAD) and Exome Aggregation Consortium (ExAC). Then, the variants were compared by multiple databases, including the Human Gene Mutation Database, OMIM, DECIPHER and ClinVar database. Only those with allele frequency ≤ 1% would be retained. To predict the pathogenicity of the mutations, several softwares were used, including the Sorting Intolerant from Tolerant (SIFT), Mutation Taster and Splice AI. The variants were interpreted under the guideline of the ACMG. In case of variants of unknown significance, only those highly compatible with the phenotype or frequently identified in more than one family would be collected and analyzed. Then, Sanger sequencing was carried out to validate the variants. The products were sequenced with the ABI 3730 DNA analyzer (Applied Biosystems™) and analyzed by DNASTAR 5.0 software.

Results

Clinical characteristics of participants

A total of 512 fetuses with anomalies were recruited after excluding clinically relevant abnormal karyotyping or chromosomal microarray findings. 501 were singleton pregnancies and 11 were twin. The median age of the pregnant women was 30 years old (range, 22–43 years) and gestational age at the time of testing was 24 (range, 13–33 weeks). Parents opted for termination in 246 pregnancies (48.05%). 12 fetuses ended in intrauterine death (2.34%), 48 were premature birth (9.38%) and 206 were full-term birth (40.23%). Demographic characteristics and outcomes of the study were shown in Table 1.

Table 1 Demographic characteristics and outcomes of the study cohort

Distribution of the fetal anomalies and its diagnosis rates

The included fetuses manifesting with variable anomalies were grouped according to the fetal phenotype. 180 fetuses with malformations involving at least two systems were classified as multisystem anomalies. 332 fetuses manifested with single-system malformations, involving 65 in skeletal, 49 in genito-urinary, 46 in central nervous systems, 44 in increased nuchal translucency (NT), 34 in cardiac, 33 in facial, 32 in fetal hydrops, 26 in abdominal and three in chest anomalies.

As is shown in the Table 2, multisystem anomalies were frequently identified, with a diagnosis rate of approximately 17.22%. Among the single anomalies, skeletal system was the most frequently affected organ referred for prenatal WES, with a detection rate of 18.46%. The molecular diagnostic rate was relative low for fetuses with anomalies affecting facial and cardiac systems. None of pathogenic variants were identified in fetuses present with isolated increased NT or chest anomalies.

Table 2 Distribution of the fetal anomalies and its diagnosis rates

Frequently identified phenotype and its diagnosis rates

In Table 3, nine phenotype were frequently identified in fetuses, including fetal growth restriction (FGR) in 46, short femur in 39, ventricular septal defect in 24, lymphatic hygroma in 20, cleft lip and/or palate in 20, kidney cyst in 18, situs viscerum inversus in 18, varus foot in 15 and ventriculomegaly in 15 cases. Among them, short femur, ventriculomegaly, kidney cyst and cleft lip and/or palate had relative high diagnosis rates. Pathogenic variants were rarely identified in lymphatic hygroma, situs viscerum inversus and varus foot.

Table 3 Frequently identified phenotype and its diagnosis rates

Recurrent occurrence of the same phenotype in siblings

Fetus presenting similar structural anomalies with its siblings were collected in the Table 4. Most of the families failed to identify genetic causes in the previous conception. When the same fetal phenotype occurred in the second conception, the couple received invasive prenatal diagnosis. Recurrent occurred fetal phenotype included interventricular septal defect, short femur, polyhydramnios, polydactyly, encephalocele, ventriculomegaly, fetal akinesia, pleural effusion, renal cyst, hydrocephalus, lymphatic hygroma and arthrogryposis.

Seven pathogenic variants were identified, six of which were inherited from the one or both of the parents. In case 1, interventricular septal defect was recurrently observed, but KMT2D was found to be de novo. In the case 4, bowed and short bones were frequently occurred, but the COL1A1 variant was inherited from unaffected mother, with an autosomal dominant inheritance manner.

Table 4 Features of the recurrent phenotype in the pedigree

Identified candidate variants and data analysis

180 nucleotide mutations in 112 genes were identified as candidate variants. Finally, 97 diagnostic variants in 65 genes were identified in 69 fetuses after phenotype-driven strategy. Among them, 29 genes were identified in more than one fetus (Fig. 1).

Fig. 1

Frequently identified genes in more than one fetus

Variable variants in TSC2 (n = 5), FGFR3 (n = 4), MUSK (n = 4), PKD1 (n = 3), COL2A1 (n = 3), COL1A1 (n = 2), HNF1B (n = 2) and TCOF1 (n = 2) exhibited highly similar phenotype. While fetuses with variants of KMT2D (n = 2) and PTEN (n = 2) present with different anomalies (Table 5).

Table 5 Features of the frequent genes and related fetal phenotype

Discussion

This study shares the clinical data of fetal WES in 512 trios presenting with structural anomalies, with an average detection rate of 13.48%. Skeletal system was the most frequently affected organ referred for prenatal WES, and short femur and kidney cyst were the most common phenotype. WES had limited value for isolated increased NT and chest anomalies.

WES was first applied in the research arena around 2011 [20]. Since then, the capture, sequencing and analysis of gene ushered an advancement swiftly and vigorously. It is an efficient diagnostic technology to identify novel pathogenic variants based on careful and complete evaluation of correlations between genotype and phenotype. The idea of performing fetal WES was endorsed in ACMG, and it has recently been considered as a first-tier method in fetal anomalies [11, 21].

The application of WES in prenatal diagnosis can assist in determining fetal prognosis and help to relieve couples’ anxiety to some extent, however, the ambiguous interpretations of variants of unknown significance may produce new anxiety [22]. As WES can generate a huge number of variants, the identification of specific phenotype in the proband is of vital significance. Regretfully, as is shown in Table 3, fetal anomalies observed in the ultrasound or MRI are more common in short limbs, large or small head circumference, polyhydramnios or oligohydramnios and so on, which are too ambiguous to guide the interpretation. In addition, some characteristics such as skin, intelligence, movement, language, hearing, smelling, touching and so on can not present in the fetal period. Therefore, the detection rate in fetal anomalies is significantly lower than that reported in children or adults, even with a similar sequencing coverage and interpretation strategy [22–24].

The application of fetal WES should grasp the indications strictly. Undoubtedly, the results of publications are strong evidence on the value of fetal WES in the clinical use. Investigations about the detection rate of fetal WES is reported every now and then. The earliest publication can be dated to 2018 [4], Normand et al. recruited 146 couples performing fetal WES from 2012 to 2017. An overall molecular diagnostic rate of 32% was concluded, which greatly promoted the application of fetal WES. Subsequently, two large cohorts were performed in 2019 [3, 13], with a rate of 8.5% and 10% respectively, enormously different from the previous conclusions. The amount, region and race discrepancy of subjects recruited might be the primary causes. Generally, detection rates will be largely overestimated in relative small cohorts [16].

The use of fetal WES is booming, with continuously evolving sequencing and analyzing technologies. Timely collection of clinical data and an accurate evaluation of detection value in large cohort is of vital significance to guide the application. Up to now, large Chinese cohort is still limited and data in different regions are valuable.

According to the statistics uncovered in our cohort, 69 fetuses were finally diagnosed, with an average detection rate of 13.48%. As for the single anomalies, the detection rate ranged from zero to 18.46%, which are conformed with the previous large cohort studies [1, 3, 13, 14]. In brief, more variants were found in fetuses diagnosed with renal anomalies. Pathogenic variants identified in cardiac, central nervous anomalies and FGR in our center were slightly less than previous reports. None of diagnosed variants were identified in isolated increased NT and chest anomalies in our clinic.

Lymphatic hygroma, situs viscerum inversus and varus foot, isolated or coupled with other anomalies, were frequently identified in prenatal ultrasound imaging, with relative low diagnostic rates. Pathogenic variants were frequently identified in fetuses manifested with short femur, ventriculomegaly, kidney cyst and cleft lip and/or palate. The anomalies, such as ventricular septal defect and cleft lip and/or palate, can be repaired after birth and may have a good prognosis. However, pathogenic genetic causes can yield contrary outcomes. Therefore, the identification of its genetic causes is of vital significance.

The recurrent phenotype is a strong indication for prenatal WES. In the current investigation, recurrent phenotype in consecutive siblings were observed in nine pedigrees, seven of which were finally diagnosed. Three diagnosed variants were inherited in an autosomal recessive manner and two were in a X-linked inheritance. In case 1, the de novo variant of KMT2D was inherited in an autosomal dominant manner, hinting a possible occurrence of germline mosaicism. In the case 4, bowed and short bones were frequently occurred, and the COL1A1 variant was inherited from unaffected mother. It means that it might have a good pregnancy outcome. However, the pregnant woman chose to terminate the pregnancy.

In conclusion, the analysis of WES data in a retrospective cohort of 512 unselected pedigrees shows the detection rate of WES in fetal anomalies. With negative findings in karyotyping and CMA, WES can add additional 13.48% detection rate in fetuses with anomalies. It provides a more comprehensive understanding of prenatal whole exome sequencing and expands phenotypic and genotypic spectrum.