Genetic diagnosis in fetal hydronephrosis: assessment using chromosomal microarray analysis and whole-genome sequencing

Abstract

Background

Antenatal hydronephrosis (AHN) is a common urological abnormality identified during prenatal ultrasound examination, with heterogeneous underlying causes. This study aimed to investigate the genetic etiologies of AHN using chromosomal microarray analysis (CMA) and whole-genome sequencing (WGS), and to identify the relationship between changes in AHN severity and genetic abnormalities.

Methods

Fetuses with AHN underwent CMA, followed by retrospective WGS. Detection rates of chromosomal abnormalities and monogenic variants were compared among different AHN types and between fetuses with downgraded versus persistent or upgraded hydronephrosis. A variant of uncertain significance (VUS) affecting a splice site was functionally validated using a minigene-splicing assay, leading to reclassification.

Results

Chromosomal abnormalities were detected in five of 54 fetuses (9.26%) using CMA, while WGS identified monogenic variants and chromosomal abnormalities in 10 of 54 (18.52%). The detection rate was significantly higher in AHN with extrarenal system malformations than in isolated AHN or AHN with renal tract malformations (p < 0.05). Fetuses with downgraded AHN had a lower detection rate (6.25%) than those with persistent or upgraded hydronephrosis (23.68%, p = 0.249). A novel GREB1L splice-site variant (c.2557G > A) caused aberrant splicing and was reclassified as likely pathogenic.

Conclusions

WGS demonstrates higher diagnostic efficiency over CMA for AHN. Persistent or worsening AHN warrants timely genetic testing, and functional assays are critical for resolving VUS in prenatal diagnosis.

Supplementary Information

The online version contains supplementary material available at 10.1186/s40246-026-00950-x.

Introduction

Antenatal hydronephrosis (AHN), characterized by abnormal dilation of the renal pelvis and calyceal system, is among the most frequently detected urological anomalies on prenatal ultrasound, with a prevalence of up to 5% in fetuses [1]. Although more than half of cases resolve spontaneously, approximately one-third persist postnatally and may lead to renal or urinary tract impairment. These persistent cases are often associated with congenital anomalies of the kidney and urinary tract, and severe obstruction can result in increased morbidity and mortality [2, 3]. A 30-year follow-up study revealed that individuals with childhood hydronephrosis have an elevated risk of developing chronic kidney disease later in life [4], underscoring the importance of early detection and intervention to preserve renal function. Prenatal diagnosis, therefore, provides an opportunity to identify high-risk fetuses before urinary symptoms, recurrent infections, or renal impairment occur, enabling timely surveillance and proactive management.

Prenatal ultrasound is the first-line method for detecting AHN, most often identified during routine obstetric ultrasonography. Classification systems based on anterior-posterior renal pelvic diameter (APRPD) are widely used to stratify the risk of clinically significant pathology into mild, moderate, and severe categories [5]. For fetuses with severe or high-risk AHN, genetic counseling and testing are recommended to assess associated conditions, detect extrarenal manifestations, and guide clinical management. Prenatal invasive testing is routinely advised when anomalies are identified on ultrasound [6]. Pathological AHN has been linked to genetic abnormalities such as trisomy 18 and 21 and to pathogenic/likely pathogenic (P/LP) copy number variants (CNVs) [7–9]. However, data on single-gene variants remain limited.

Despite several existing guidelines and grading systems, no single finding or study can reliably distinguish all clinically significant cases from physiologic or self-resolving hydronephrosis [10–12]. Thus, clinicians must balance minimizing unnecessary testing and parental anxiety against the need to detect significant diseases or genetic abnormalities without delay.

Next-generation sequencing has emerged as a potent tool for diagnosing Mendelian diseases. Whole-genome sequencing (WGS), in particular, can identify diverse variant types, including CNVs, single-gene variants, intronic single-nucleotide variants (SNVs), SNVs in non-coding RNA, and mitochondrial DNA variants [13]. In this study, we applied chromosomal microarray analysis (CMA) and WGS to 54 fetuses with AHN to investigate underlying genetic causes and evaluate the clinical utility of WGS. By longitudinally monitoring fetal hydronephrosis, we further explored the relationship between hydronephrosis and genetic findings, aiming to enhance prenatal diagnostic decision-making.

Methods

Subjects

The design and protocol of this study were reviewed and approved by the ethics committee of Changzhou Maternity and Child Health Care Hospital (No. 202441). All pregnant women received genetic counseling and provided informed consent prior to testing.

Between January 2020 and December 2023, 54 pregnant women with AHN identified through systematic prenatal ultrasonography were referred to three prenatal diagnostic centers in Changzhou, Suzhou, and Lianyungang for invasive genetic testing. After comprehensive genetic counselling, all patients underwent amniocentesis for confirmatory diagnosis using CMA. Residual DNA samples were archived and subsequently used for retrospective WGS (Fig. 1).

Fig. 1.

Schematic of the study design. AHN antenatal hydronephrosis, CMA chromosomal microarray analysis, WGS whole-genome sequencing, P pathogenic, LP likely pathogenic, VUS variants of uncertain significance, CNVS copy number variants, SNVs single-nucleotide variants

Cases were classified into three groups based on antenatal findings [14]: (i) isolated AHN without additional abnormalities, (ii) AHN with other renal tract abnormalities (e.g., hydroureter) but normal extrarenal findings, and (iii) AHN with concomitant extrarenal abnormalities. AHN severity was graded according to the APRPD as follows: mild, 4–6 mm before 28 weeks of gestation and 7–9 mm after 28 weeks, moderate, 7–10 mm before 28 weeks and 10–15 mm after 28 weeks, and severe, > 10 mm before 28 weeks and > 15 mm after 28 weeks. After the initial detection of AHN, each patient underwent serial prenatal ultrasounds at least three times, with intervals of 2–4 weeks between each examination. APRPD measurements were used to evaluate severity and progression, and serial changes were categorized as downgraded or upgraded/persistent.

Chromosomal microarray analysis

The procedure for prenatal CMA has been described previously [15]. Amniotic fluid samples were collected, and genomic DNA was extracted using a QIAamp DNA Mini Kit (Qiagen Inc., Valencia, CA, USA). DNA (250 ng) was amplified, labeled, and hybridized to a GCS 3000Dx v.2 microarray platform (Affymetrix, CA, USA). The SNP array test was performed using an Affymetrix CytoScan 750 K Array. Data retrieval was performed using public databases, including DECIPHER, OMIM, ClinVar, ISCA, NCBI, and the UCSC. The pathogenicity of the identified CNVs was evaluated according to the guidelines of the American College of Medical Genetics and Genomics (ACMG).

Whole-genome sequencing

A total of 300 ng of genomic DNA was enzymatically fragmented using a shearing enzyme premix (ENZYMATICS) to generate fragments with a primary size distribution of 100–500 bp. Size selection was performed with magnetic beads (VAHTS™ DNA CLEAN BEADS, VAZYME) to enrich fragments between 200 and 300 bp. Fragmented DNA then underwent end repair, adenylation at the 3’ end, and ligation with specialized adapters containing a complementary thymine overhang. The adapter-ligated DNA library was purified with magnetic beads, followed by pre-polymerase chain reaction (PCR) amplification and purification to complete the library construction. Library concentration and quality were assessed before single-strand circularisation. The circularized libraries underwent rolling circle amplification to form DNA nanoballs, which were pooled into a mixed library and loaded onto a sequencing chip. High-throughput sequencing was performed on a DNBSEQ-T7 sequencer using a PE100 + 10 sequencing strategy, generating at least 120 GB of raw sequencing data per sample.

Bioinformatic analysis was initiated upon obtaining the raw sequencing data. SOAPnuke [16] was used for quality control to remove low-quality reads and adaptor contamination, yielding clean reads. The clean reads were aligned to the hg19 (GRCh37) reference genome (https://www.ncbi.nlm.nih.gov/assembly/GCF_000001405.13/) using the Burrows-Wheeler Aligner [17]. SNV and insertions/deletions (Indels) were identified using the Genome Analysis Toolkit [18]. The identified variants were annotated and filtered by comparison with multiple databases, including NCBI dbSNP v147, dbNSFP v2.9.1, ESP6500 v2, HapMap, the 1000 Genomes Project, and a dataset of 100 healthy Chinese adults. Additional annotation and filtering were conducted using BGI’s proprietary BGI-Varanno algorithm. CNVs were detected using CNVnator [19], and structural variants were identified using LUMPY [20]. Variant pathogenicity was interpreted according to the ACMG guidelines [21], with references to ClinVar (https://www.ncbi.nlm.nih.gov/clinvar/), OMIM (https://www.omim.org/), and HGMD (http://www.hgmd.cf.ac.uk/). ExpansionHunter [22] was used to detect dynamic mutations within the genome.

Sanger sequencing

Primers were designed upstream and downstream of the detected variant sites. The target regions were amplified by PCR, and the PCR products were subjected to Sanger sequencing. The sequencing results were compared with the corresponding reference gene sequences to validate their accuracy.

Splicing prediction and functional validation

The potential effects of variants on splicing were predicted using three splice-site prediction tools: SpliceAI (https://spliceailookup.broadinstitute.org/), RDDC (https://rddc.tsinghua-gd.org/), and FF (https://www.fruitfly.org/seq_tools/splice.html). Functional validation of a noncanonical splice site variant (c.2557G > A) in the GREB1L gene was performed using a minigene splicing assay. Amplicons were generated through overlapping PCR and restriction enzyme digestion and cloned into two vector systems: pcMINI-GREB1L and pcMINI-N-GREB1L. PCR amplification followed by Sanger sequencing confirmed successful construction of both wild-type (WT) and mutant (Mut) expression vectors. Recombinant plasmids (pcMINI-GREB1L-wt/mut and pcMINI-N-GREB1L-wt/mut) were transiently transfected into human embryonic kidney (HEK-293T) and cervical cancer (HeLa) cells following the manufacturer’s guidelines. After 48 h, total RNA was extracted using the TRIZOL method (RNA extraction kit, Vazyme), and complementary DNA (cDNA) was synthesized using a reverse transcription kit (Takara) according to the manufacturer’s instructions. The PCR-amplified products were analysed by 1.5% agarose gel electrophoresis and verified by Sanger sequencing.

All primers used in this study are listed in Supplementary Table S1, and the construction and verification of recombinant plasmids (pcMINI-GREB1L-wt/mut and pcMINI-N-GREB1L-wt/mut) are illustrated in Fig. 2A and Supplementary Fig S1A.

Fig. 2.

Minigene study on a noncanonical splice site variant in GREB1L. A Schematic illustration of cloned vectors. Red*indicates the mutation position. B Sequencing results of the target fragment with wild-type (wt) at the top and mutant (mut) at the bottom. C Reverse-transcription polymerase chain reaction (RT-PCR) product were separated by electrophoresis in HEK-293 T (left) and HeLa (right) cells. D Minigene product sequencing results: a, normal mRNA composed of exon 18; b, skipping of exon 18 (134 bp deletion); c, retaining a 104-bp segment from the 3’ end of intro 17 downstream of exon 18

Follow-up study

Follow-up data were independently collected by two clinicians via telephone interviews, the hospital’s electronic information system, and patient files.

Statistical analysis

Categorical variables were compared using the chi-square test or Fisher’s exact test with SPSS software version 26.0. All tests were two-tailed, and statistical significance was defined as P < 0.05.

Results

Comparison of the detection rates of WGS and CMA

Among 54 pregnancies with AHN that underwent both CMA and retrospective WGS, the mean maternal age was 29.3 years (range: 21–44), and the mean gestational age at ultrasound detection was 25 weeks (range: 18.5–30). CMA identified six P/LP chromosomal aberrations in five fetuses (9.26%), including one aneuploidy and five P/LP CNVs. Retrospective WGS detected P/LP variants in ten fetuses (18.52%); five fetuses with six chromosomal abnormalities were also detected by CMA, and five fetuses with additional sequence variants were undetectable by CMA, yielding an extra diagnostic yield of 9.26% (5/54) (Table 1).

Table 1.

Detailed information on genetic abnormalities identified by CMA or WGS in 10 fetuses with AHN

case	GA (w)	Ultrasonic findings	APRPD (mm)	CMA	WGS	Disease	Pregnancy outcome
1	20.4	Right hydronephrosis	R:10	arr[GRCh37] 22q11.21(21059669_21460471)x1	chr22:21044001–21466000:loss	22q11.2(LCRC-D) deletion syndrome	TOP
2	24	Bilateral hydronephrosis, Cleft lip and palate	L:9.5 R:16	arr[GRCh37] 7q11.23(72347978_76712260)x3	chr7:72348001–767326000:gain	Williams-Beuren syndrome	TOP
3	22	Bilateral hydronephrosis	L:6.6 R:7.8	arr[GRCh37] 15q11.2(22770422_23191761)x1	chr15:22832001–23315000:loss	15q11.2 deletion syndrome	Live birth
4	25.1	Bilateral hydronephrosis, intrauterine growth retardation, VSD	L:8.8 R:5.9	arr(9)x2–3	chr9: mosaic gain	trisomy 9 syndrome	TOP
5	23.4	Cleft lip and palate, abnormal right thumb, VSD, right kidney hydronephrosis and enlargement	R:10	arr[GRCh37] 4q34.3q35.2(179774679_190957460)x1 arr[GRCh37] 1q31.2q44(193388230_249224684)x3	chr4:179673001–190943000:loss chr1:193321001–249273000:gain	partial trisomy 1q31 syndrome ; 4q35 deletion	TOP
6	18.3	Large bladder, mild hydronephrosis, fetal ascites, cystic mass in abdominal cavity	L:5.3 R:6.4	Normal	P ACTG2 NM_001615.3:c.118 C > T(p.Arg40Cys)	Megacystis-Microcolon-Intestinal Hypoperistalsis Syndrome	IUFD
7	24.3	Bilateral hydronephrosis, prestenotic narrowing in the upper urinary tract	L:15 R:5.3	Normal	LP BRAF NM_004333.4:c.1805 C > T(p.Ser602Phe)	Noonan syndrome	Premature birth
8	24	Bilateral hydronephrosis	L:21 R:11	Normal	LP RRAS2 NM_001177314.1:c.110 A > G(p.Gln37Arg)	Noonan syndrome	TOP
9	24	Right hydronephrosis, PL-SVC	R:11	Normal	LP NRAS NM_002524.4:c.193 A > G(p.Ser65Gly)	Noonan syndrome	Live birth
10	24	Left hydronephrosis	L:4.2	Normal	LP GREB1L NM_00114296 6.1:c.2557G > A	Renal Hypodysplasia/Aplasia 3	Live birth

GA gestational age, APRPD anterior-posterior renal pelvic diameter, CMA chromosomal microarray analysis, WGS whole-genome sequencing, CNV copy number variant, P pathogenic, LP likely pathogenic, TOP termination of pregnancy, IUFD intrauterine fetal demise, VSD ventricular septal defect, PL-SVC Persistent Left Superior Vena Cava, L left, R right

Pathogenic genetic abnormalities were detected in 18.52% of fetuses with AHN, comprising one aneuploidy (10%), five CNVs in four fetuses (40%), and five single-gene variants (50%). In one fetus with mosaic trisomy 9, both WGS and CMA confirmed a 40% mosaic level. The P/LP CNVs consistently identified by both methods included a 22q11.21 deletion, 7q11.23 duplication, 15q11.2 deletion, partial trisomy 1q31, and 4q35.2 deletion, ranging from 480 kb to 55.8 Mb (Table 1).

Of the P/LP single-gene variants, four were missense mutations and one was a frameshift mutation. Missense variants in BRAF, RRAS2, and NRAS were potentially associated with Noonan syndrome, while a missense ACTG2 variant occurred in a fetus with extra-renal malformations, possibly related to megacystis-microcolon-intestinal hypoperistalsis syndrome. A splice-site variant in GREB1L may cause frameshift mutations linked to renal hypodysplasia/aplasia-3 (RHDA3) (Table 1).

Diagnostic rate analysis of different subgroups

Across all phenotypic subgroups, WGS achieved a higher diagnostic yield than CMA. In isolated AHN (n = 27), WGS identified P/LP variants in four cases (14.81%) versus two (7.41%) by CMA. Among the 20 cases with additional renal tract malformations, WGS identified one P/LP single-gene variant, while CMA detected none. In fetuses with extrarenal malformations (n = 7), WGS detected genetic abnormalities in five cases (71.43%) and CMA in three (42.86%). The diagnostic rate in this subgroup was significantly higher than in the isolated AHN and renal tract malformation groups (p = 0.001; Table 2). Longitudinal ultrasound follow-up showed improvement in hydronephrosis (downgrade group; in 16 of the 54 fetuses [29.63%]), comprising eight with isolated AHN, six with renal tract malformations, and two with extrarenal anomalies. Hydronephrosis persisted or worsened (upgrade/persist group; 38 fetuses [70.37%]), including 19 with isolated AHN, 14 with renal tract malformations, and five with extrarenal anomalies. The downgraded group harbored only one P/LP variant and six variants of uncertain significance (VUS), with no pathogenic CNVs identified by either WGS or CMA. The upgrade/persist group had four P/LP single-gene variants and five P/LP CNVs detected by WGS, plus 11 VUS. The detection rate of genetic abnormalities was 6.25% in the downgraded group and 23.68% in the upgrade/persist group, though the difference was not statistically significant (p = 0.249; Table 3).

Table 2.

Prenatal diagnosis and follow-up results of fetuses with AHN

Group	n	Total genetic abnormalities n(%)	CMA		WGS		Outcome
			P/LP n (%)	VUS n (%)	P/LP n (%)	VUS n (%)	IUFD	TOP	Live Birth	Surgical intervention
Isolated AHN	27	4 (14.81%)	2 (7.41%)	1(3.70%)	4 (14.81%)	8 (29.63%)	0	5	20	2
AHN with renal tract malformation	20	1 (5.00%)	0 (0)	1 (5.00%)	1 (5.00%)	9 (45.00%)	0	2	14	4
AHN with extrarenal system malformation	7	5* (71.43%)	3 (42.86%)	0 (0)	5 (71.43%)	2 (28.57%)	1	3	3	0
Total	54	10 (18.52%)	5 (9.26%)	2 (3.70%)	10 (18.52%)	17 (31.48%)	1	10	37	6

AHN antenatal hydronephrosis, CMA chromosomal microarray analysis, WGS whole-genome sequencing, P pathogenic, LP likely pathogenic, VUS variants of uncertain significance, TOP termination of pregnancy, IUFD intrauterine fetal demise

*Total genetic abnormalities in the AHN group with extrarenal system malformation versus the isolated AHN and AHN groups with renal tract malformation: p = 0.001, Fisher’s exact test

Table 3.

CMA and WGS results of fetuses with AHN between the downgraded and upgraded/persistent groups

Group	n	Clinical information			CMA		WGS
		Isolated AHN n (%)	AHN with renal tract malformation n (%)	AHN with extrarenal system malformation n (%)	P/LP n (%)	VUS n (%)	P/LP n (%)	VUS n (%)
Downgrade	16	8 (50.00%)	6 (37.50%)	2 (12.50%)	0 (0/16)	0 (0/16)	1 (6.25%)	6 (37.50%)
Upgrade/persistent	38	19 (50.00%)	14 (36.84%)	5 (13.16%)	5 (13.16%)	2 (5.26%)	9 (23.68%)	11 (28.95%)
Total	54	28 (51.85%)	20 (37.04%)	6 (11.11%)	5 (9.26%)	2 (3.70%)	10 (18.52%)	17 (31.48%)

Total genetic abnormalities in the downgrade versus upgrade/stent group: p = 0.249, Fisher’s exact test

Minigene study on the noncanonical splice site variant in GREB1L

In Case 10, mild bilateral hydronephrosis was detected via ultrasound at 24 weeks of gestation. Subsequent cardiac ultrasound revealed left atrial enlargement and patent ductus arteriosus. The mild bilateral hydronephrosis persisted to 35 weeks, with progression to moderate on the left side until delivery. WGS identified the GREB1L variant (NM_001142966.1: c.2557G > A), initially classified as VUS, based on PM2_Sup and PP3_Str.

Based on deleterious predictions from multiple in silico tools, we validated the potential aberrant splicing effect of the variant using a minigene splicing assay, which recapitulates the in vivo splicing process [23]. WT and Mut pcMINI-GREB1L /pcMINI-N-GREB1L plasmids were constructed and transfected into HeLa and HEK293T cells. After 48 h, total RNA was extracted and reverse-transcribed into cDNA.

The target region encompassing GREB1L exon 18 was amplified by PCR, and the products were analyzed by agarose gel electrophoresis. The Mut construct yielded three bands (A, B, and C), whereas the WT produced a single band (A). Sequencing and gel analyses revealed that Mut Band A retained exon 18, mirroring the WT transcript. Band B lacked exon 18 (134 bp deletion), and C retained a 104-bp segment from the 3’ end of intron 17 downstream of exon 18 (Fig. 2C, D). Based on this functional evidence and updated interpretation criteria, the variant was reclassified as LP, supported by PM2_Sup and PVS1.

Discussion

Detection capability and additional findings of WGS

This study compared the diagnostic performance of CMA and WGS in fetuses with AHN, integrating detailed ultrasound findings for grading and correlation analysis. CMA identified chromosomal abnormalities in five of 54 fetuses (9.26%), whereas WGS detected monogenic variants and chromosomal abnormalities in ten of 54 fetuses (18.52%). This represents an incremental diagnostic yield of 9.26% for WGS.

Pathogenic genetic abnormalities were detected in 18.52% of fetuses with AHN: aneuploidies accounted for 10%, CNVs for 40%, and single-gene variants for 50%. As a unified testing platform, WGS can simultaneously identify a broad range of genomic alterations, including aneuploidies, SNVs/indels, CNVs, and mitochondrial variants [24]. Given its increasing diagnostic yield and steadily declining costs, WGS is now applied across multiple clinical fields to investigate genetic etiologies and facilitate diagnosis. Numerous studies have confirmed its superiority compared to traditional genetic testing, supporting its adoption as a first-line diagnostic tool for rare genetic disorders and pediatric diseases [24, 25].

Among the pathogenic single-gene variations detected by WGS, three were associated with Noonan syndrome. As shown in Table 1, cases 7–9 harbored missense mutations in BRAF, RRAS2, and NRAS, respectively. Noonan syndrome is a multisystemic genetic disorder characterized by a range of physical features and developmental abnormalities. Prenatally, affected fetuses may present with increased nuchal translucency, cystic hygroma, fetal hydrops, congenital heart defects, hydronephrosis, and polyhydramnios [26]. In our study, three fetuses with hydronephrosis were diagnosed with Noonan syndrome. Prenatal ultrasound in all three revealed severe unilateral or bilateral hydronephrosis. One case was associated with ureteral stenosis, another exhibited persistent left superior vena cava. The latter case showed a reduction of hydronephrosis to a moderate level after three weeks of monitoring, while cardiac anomalies persisted throughout gestation. The third fetus presented with isolated hydronephrosis, and fetal echocardiography during the late second trimester revealed mild tricuspid regurgitation and fossa ovalis dilation.

Although the sample size was limited, the occurrence of Noonan syndrome in fetuses with AHN suggests a possible clinical association. Recent evidence supports this, showing that renal abnormalities occur more frequently in Noonan syndrome than previously recognized (31% vs. 10%), with hydronephrosis being the most prevalent manifestation, affecting approximately 70% of those with renal involvement [27].

Clinical and subgroup analysis implications

In subgroup analysis, fetuses with AHN accompanied by extrarenal malformations exhibited a higher prevalence of genetic abnormalities (71.43%). Mosaic aneuploidy and CNVs accounted for 42.86% of cases, while single-gene variants comprised 28.57%. This elevated genetic burden may reflect pleiotropic gene effects, dosage sensitivity of key developmental genes, and disruption of coordinated regulatory networks, affecting gene regulation, interactions, and pathway dysregulation [28–30].

All cases underwent longitudinal follow-up of hydronephrosis throughout pregnancy. Previous studies indicate that approximately one-third of AHN cases resolve spontaneously before birth, another third within the first few years of life. Transient or nonobstructive dilatation of the pelvi-calyceal system is commonly seen among spontaneously resolving cases [3, 31, 32]. Clinical guidelines recommend serial ultrasonographic surveillance (every two to three weeks) for fetuses with bilateral APRPD > 4 mm or those with a solitary affected kidney and normal amniotic fluid volume in the second trimester. In our study, among 16 fetuses with downgraded hydronephrosis, one P/LP variant (6.25%) was identified, compared to four P/LP variants and five CNVs in the persistent /upgraded group (23.68%). Although the difference was not statistically significant (p = 0.249), likely due to the limited sample size, a clear trend emerged toward higher pathogenic genetic abnormalities in the persistent/upgraded group. These findings suggest that fetuses with persistent or progressive hydronephrosis are more likely to harbor clinically significant genetic abnormalities and should be prioritized for prenatal genomic evaluations. Furthermore, accumulating evidence indicates that timely management is required for optimal outcomes in patients with persistent or worsening AHN [32, 33].

The role of the minigene study on the variant in GREB1L

In case 10, a GREB1L variant was initially categorized as a VUS but later reclassified as LP following in vitro functional validation. This mutation, located within the initial sequence of exon 18, represents a non-classical splicing alteration not previously reported. Minigene splicing assays revealed that, in addition to the normal transcript, the Mut variant induced two aberrant transcripts: one skipping exon 18 and another retaining a 104-bp segment from intron 17. Both aberrant transcripts introduced premature termination codons, resulting in truncated proteins of 871 and 884 amino acids, respectively.

These results provide functional evidence for the pathogenicity of this novel GREB1L variant, which may be closely associated with RHDA3, a disorder with wide phenotypic variability, ranging from hydronephrosis and vesicoureteral reflux to unilateral renal hypoplasia, horseshoe kidney, or unicornuate uterus. Some carriers remain asymptomatic, suggesting incomplete penetrance. In case 10, unilateral mild hydronephrosis was initially observed, followed by cardiac abnormalities, including left atrial enlargement and patent ductus arteriosus. This aligns with emerging data linking GREB1L variants to congenital heart defects, including aortic stenosis, left ventricular hypertrophy, and outflow tract anomalies [23, 34, 35]. This case underscores the critical role of in vitro functional assays in refining variant interpretation. Incorporating minigene-based splicing validation enabled reclassification of the GREB1L VUS to LP, underscoring the importance of experimental evidence in variant assessment, particularly for atypical splice-site alterations.

According to ACMG guidelines, clinical reporting of genetic findings during pregnancy should focus on P and LP variants, while VUS should be reported only when strongly associated with the observed phenotype [36]. Studies show that approximately 5% of prenatal VUS are later reclassified with the emergence of additional phenotypic data [37], and one-third of cases may be upgraded to LP when postnatal findings are integrated [38]. Incorporating in vitro functional analyses during pregnancy can further clarify uncertain findings, support informed counseling, optimize prenatal management, and early therapeutic interventions.

Limitations

This study had certain limitations. There is currently no standardized prenatal diagnostic protocol for AHN, and the potential risks of invasive testing often lead pregnant women to decline or defer such procedures, limiting sample size and statistical power. The genetic etiology of AHN appears heterogenous, encompassing aneuploidy, CNVs, and a substantial proportion of single-gene variants; however, our modest cohort size precluded precise estimation of variant frequencies.

While WGS demonstrates strong diagnostic potential, several challenges remain. Accurate variant interpretation relies heavily on phenotypic data, which are constrained in prenatal settings. Fetal imaging has limited sensitivity for subtle dysmorphic or neurodevelopmental anomalies, and some phenotypes evolve with gestational age, further complicating prenatal characterization. Furthermore, comprehensive genome-wide analysis increases the likelihood of identifying VUS [39], posing clinical challenges given the limited timeframe for functional follow-up studies during pregnancy.

Conclusions

This study evaluated fetuses with AHN who underwent CMA followed by retrospective WGS. WGS provided a significant improvement in prenatal genetic diagnosis, yielding an additional 9.26% detection rate beyond CMA by identifying pathogenic single-nucleotide variants undetectable by microarray. Functional validation of VUS, particularly those in genes associated with observed phenotypes, is essential for accurate interpretation. Furthermore, persistent or progressive hydronephrosis was associated with a higher genetic burden, highlighting the importance of comprehensive genomic evaluation for guiding prenatal counseling and clinical decision-making.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

JW and BY designed the study. WW and XZ collected the samples. JW and JZ conducted genetic analyses. JW, XZ and BZ collected clinical information. JW and JZ drafted the manuscript. JW, BZ, WW and XZ provided funding acquisition.

Funding

This study was funded by the Nanjing Medical University Changzhou Medical Center Project(CMCC202421), the Top Talent of Changzhou “The 14th Five‑Year Plan” High‑Level Health Talents Training Project (2022CZBJ089), Major Science and Technology Project of Changzhou Municipal Health Commission (ZD202418), Medical Technology Research Project of Changzhou Maternal and Child Health Care Hospital (YJ202413,YJ202512).

Data availability

The datasets presented in this article are not readily available because Regulations on the management of human genetic resources in China. Requests to access the datasets should be directed to the corresponding author.

Declarations

Ethics approval and consent to participate

The study design and protocol were reviewed and approved by the Ethics Committee of Changzhou Maternity and Child Health Care Hospital (No. 202441) and were conducted in accordance with the principles of the Declaration of Helsinki. All pregnant women received genetic counseling and provided informed consent prior to testing.

Competing interests

The authors declare no competing interests.