Evaluating the effectiveness of routine noninvasive prenatal screening for CNVs in 22q11.2 region in a cohort of 38,495 pregnancies

Abstract

22q11.2 deletion syndrome (22q11.2 DS) is the second most common cause of congenital heart disease. The American College of Medical Genetics and Genomics (ACMG) has recently recommended implementing non-invasive prenatal screening (NIPS) for 22q11.2 DS for all pregnant women. This study aims to assess the effectiveness of routine NIPS for screening 22q11.2 deletion in a cohort of 38,495 pregnancies from the general population. We conducted a retrospective analysis of 38,495 pregnant women who underwent NIPS at Longgang Maternal and Child Health Hospital in Shenzhen from December 2022 to March 2024. Chromosomal microarray analysis (CMA) was performed on fetuses and pregnant women identified as high-risk for 22q11.2 DS by NIPS, using amniotic fluid samples and the leukocyte cells, respectively. Of the 38,495 cases, 22 were identified as high risk for 22q11.2 deletion by NIPS. Of these, 17 underwent amniocentesis, and 5 refused prenatal diagnosis. Concordant results between CMA and NIPS were observed in 8 cases, giving a positive predictive value (PPV) of 47.06% (8/17). Nine fetuses did not show 22q11.2 deletion by CMA, although the mothers of three fetuses were identified as having a maternal 22q11.2 deletion. Follow-up on 5 cases without prenatal diagnosis revealed one postpartum case of congenital heart disease (ventricular septal defect and atrial septal defect), three cases that were terminated, and one case that continued with normal fetal development. Routine non-invasive prenatal screening for 22q11.2 deletion demonstrates practical clinical utility and provides valuable insights for identifying pregnant women at risk for 22q11.2DS.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-025-33979-4.

Introduction

22q11.2 deletion syndrome (22q11.2 DS, OMIM 611867) is the most common cause of DiGeorge syndrome (OMIM 188400), it is also causes Velocardiofacial syndrome, Conotruncal anomaly face syndrome, autosomal dominant Opitz G/BBB syndrome, Sedlackova syndrome and Cayler cardiofacial syndrome¹. 22q11.2 DS is a common chromosomal deletion disorder with an estimated prevalence of 1 in 990 to 1 in 4000 live births, 1 in 1497 miscarriages, and up to 1 in 800 to 1 in 1000 fetuses^2–7. This syndrome presents with a range of clinical features, including congenital heart disease (particularly conotruncal malformations), facial anomalies, immunodeficiency, palatal anomalies, developmental delay, and learning disabilities^8–10. Early prenatal detection is essential for effective genetic counseling, timely intervention, and management.

While congenital heart defects are the most common ultrasound finding in 22q11.2 DS, fetal anomalies are often detected in the second trimester or later. Therefore, relying on ultrasound alone may lead to missed diagnoses, especially in cases with atypical presentations. Effective prenatal screening is essential to provide comprehensive pregnancy management and intervention planning information.

Non-invasive prenatal screening (NIPS) is highly effective in detecting trisomy 21, 18, and 13 due to its high sensitivity and specificity^11,12. Advances in screening technology, including improved algorithms, increased sequencing depth, and improved cell-free fetal DNA analysis, have expanded NIPS to include screening for sex chromosome aneuploidy (SCA) and copy number variation (CNV)^12–14. High detection rates and positive predictive values (PPVs) have been reported for several microdeletion and microduplication syndromes, including DiGeorge syndrome, Prader-Willi syndrome, and Angelman syndrome^15,16.

The American College of Medical Genetics and Genomics (ACMG) recently recommended NIPS for 22q11.2 DS in all patient populations¹⁷. Despite existing studies on the efficacy of NIPS for 22q11.2 deletion screening^15,18–20, further validation in large, general populations is needed. This study aims to assess the efficacy of routine NIPS in screening for 22q11.2 deletion in a cohort of 38,495 pregnancies and to demonstrate its utility in routine screening for this syndrome.

Materials and methods

Subjects

From November 2022 to March 2024, NIPS screening was performed on 38,495 pregnant women aged 20 to 41 years with gestational ages ranging from 12 to 28^+ 6 weeks at the Shenzhen Longgang District Maternal and Child Healthcare Hospital. The study was conducted by the Declaration of Helsinki of 2013 and approved by the Medical Ethics Committee of Shenzhen Longgang District Maternal and Child Health Hospital (No. LGFYKYXMLL-2024-39). Written informed consent was obtained from all participants before testing.

Extraction, enrichment library construction, and sequencing

Cell-free fetal DNA(cffDNA) was extracted from 5 mL peripheral blood samples using a DNA extraction kit (BGI, Wuhan, China). The sample was centrifuged at 16,000g for 10 min at 4 °C. To enrich for larger cffDNA fragments, 32 µL of cffDNA solution was mixed with 39 µL of AMPureXP magnetic beads (Beckman Coulter, USA). An additional 30 µL of AMPureXP magnetic beads were added to capture shorter cffDNA fragments. After two washes with 75% ethanol, 42 µL of elution buffer was used to elute the DNA, which was then collected.

The DNA libraries were prepared for sequencing by first denaturing the DNA into single strands. DNA nanoballs (DNB) polymerase was added for rolling amplification. The DNBs were then loaded onto a chip, which was incubated before starting the sequencing process. Sequencing was performed on the MGIseq2000 system (BGI).

Bioinformatic analysis

Bioinformatic analysis was performed using Halos software (BGI), with GRCh37 as the reference genome. A z-score cut-off of 3 was used to identify trisomy 13, 18, and 21. Fetal CNVs were detected using the Hidden Markov Model (R package HMM, v1.0.1). The fetal fraction was calculated using Halos software, and results with less than 3.5% were classified as test failures.

Algorithms for CNV analysis

The human reference genome (hg19) was divided into 50 base pairs (bp) simulated reads for alignment. Raw reads from each sample were mapped to hg19, and the corrected read number (CRN) for each window was calculated using the Z-score formula.

where CRN_win is the CRN number of a given window, CRN_mean is the mean CRN across all samples, and SD_CRN is the standard deviation.

Amniotic fluid collection and DNA extraction

Amniocentesis was performed in high-risk cases, and 5 mL of amniotic fluid was collected. DNA was extracted from the amniotic fluid using the QIAamp DNA Blood Mini Kit (No. 51106, Qiagen, Germany) according to the manufacturer’s instructions.

Chromosomal microarray analysis (CMA)

CMA was performed using the Affymetrix Cytoscan 750 K chip (Thermo Fisher, CA, USA). DNA was prepared according to the manufacturer’s protocol, including digestion, ligation, PCR amplification, purification, fragmentation, and labeling for hybridization. The chip was scanned, and data were analyzed using Chromosome Analysis Suite (ChAS) 4.3 software. CNV results were filtered using a threshold of 100 kilobase pairs (Kb) and 50 probe markers. Databases such as UCSC Genome Browser, Database of Imbalance and Phenotype in Humans using Ensembl Resources (DECIPHER), Database of Genomic Variants (DGV), and Clinical Genome Resource (ClinGen) were consulted for CNV classification and interpretation²¹.

Multiplex ligation-dependent probe amplification (MLPA) analysis

MLPA was performed to validate the genetic origin of the 22q11.2 CNVs using the P064 reagent (MRC, Holland), which contains 9 probes to detect 22q11.2 deletions and duplications. The procedure was performed according to the manufacturer’s instructions, and the data were analyzed using Coffalyser.Net software (v. 220513).

Karyotyping analysis

Amniotic fluid cells were aseptically inoculated into two separate cell culture flasks under aseptic operating conditions. Chromosome preparations were made using standard cytogenetic techniques. Karyotyping was performed using an automated Leica system, and only well-distributed banded clones with chromosomes showing no or relatively few (≤ 5) crossovers were selected for subsequent analysis.

Pregnancy outcome follow-up

Pregnancy outcomes were monitored in all high-risk NIPS cases and in those with confirmed 22q11.2 DS. Follow-up extended up to 3 months post-birth.

Statistics

Statistical analyses were conducted using SPSS software version 20.0 (SPSS, Inc., Chicago, IL). Data were expressed as standard deviation, and comparisons were made using the chi-squared test. A p < 0.05 was considered statistically significant.

Results

Aneuploidy and CNV screened by NIPS

Between November 2022 and March 2024, NIPS was performed in 38,495 pregnant women. In 16 cases, the test failed due to insufficient fetal DNA concentration (Fig. 1). Among the remaining cases, the screening identified 54 positive for trisomy 21(T21), 15 for T18, 8 for T13, 167 for SCA, and 154 for CNVs (Fig. 1, Supplementary Table 1). Among the 154 CNV-positive cases, 22 were identified as high risk for 22q11.2 deletion, corresponding to an incidence rate of 0.057% (1/1750). Four cases were identified as high risk for 22q11.2 duplication (Fig. 2, Supplementary Table 2). The remaining 38,081 cases were classified as low risk.

Fig. 1.

Flowchart of routine NIPS for 22q11.2DS in 38,495 pregnant women. This figure illustrates the process and results of non-invasive prenatal screening (NIPS) for 22q11.2 deletion syndrome (22q11.2DS) across a general population of 38,495 pregnant women. It includes details of test results, confirmation by chromosomal microarray analysis (CMA), and subsequent management.

Fig. 2.

NIPS high-risk screening cases. (A) NIPS reveals a 1.72 Mb microdeletion in the 22q11.2 region, indicated by red dots with reduced sequencing results, suggesting a microdeletion. (B–D) NIPS identifies microdeletions of 1.59 Mb, 2.13 Mb, and 2.03 Mb in the 22q11.2 region. (E) The pink bar indicates a microdeletion of 2.89 Mb, presumably to be of maternal origin. (F) The pink bar shows a 3.09 Mb microduplication, also of maternal origin.

Characteristics of high-risk cases for 22q11.2 deletion

The 22 high-risk cases for 22q11.2 deletion had a median maternal age of 29.92 years and a median gestational age of 16^+ 2 weeks. The median concentration of cell-free fetal DNA in these cases was 16.42%, with an average of 8.34 million unique sequencing reads. These indicators were not statistically significantly different from low-risk cases (Table 1). Despite enrichment efforts, 5.27% of cases had less than 10% cffDNA concentration (Supplementary Fig. 1).

Table 1.

Difference analysis between 22q11.2DS high-risk and low-risk populations.

	22q11.2DS high-risk	22q11.2DS low-risk	p value
Age (years)	29.91 (SD = 5.13)	29.53 (SD = 4.57)	0.7317
Pregnant days	115.32 (SD = 12.74)	116.51 (SD = 21.75)	0.6646
Unique readers(M)	8.34 (SD = 2.29)	7.90 (SD = 1.89)	0.3846
cffDNA (%)	16.42 (SD = 4.96)	18.54 (SD = 5.86)	0.0639

*cffDNA, cell-free fetal DNA; SD, Standard deviation.

Among the 22 cases with a 22q11.2 deletion, the median deletion size was 1.79 megabase pairs (Mb). Three cases (cases 13, 14, and 16) were suspected to have CNVs of maternal origin, with a median CNV size of 3.12 Mb (Table 2). The remaining 18 cases had a median CNV size of 1.58 Mb, with 15 cases having smaller than 2 Mb (ranging from 1.04 Mb to 1.92 Mb) (Table 2).

Table 2.

Detailed information of 22 cases with NIPS 22q11.2DS high-risk screening.

Case no.	Age	GA (week)	cffDNA (%)	NIPS results	NIPS breakpoints	CMA results	CNV size	Region	Genetic origin	Outcome
1	34	19+ 2	16.68	del(22q11.21,1.92 M)	18,578,067–20,501,581	arr[GRCh37] 22q11.21(18,631,365_21,800,471)x1	3169 Kb	LCR22-A-D	de novo	TOP
2	31	16	13.54	del(22q11.21,2.13 M)	18,375,614–20,501,581	arr[GRCh37] 22q11.21(18,648,856_21,800,471)x1	3152 Kb	LCR22-A-D	de novo	TOP
3	31	17+ 3	17.44	del(22q11.21,1.2 M)	19,061,538–20,265,139	arr[GRCh37] 22q11.21(18,648,856_21,800,471)x1	3152 Kb	LCR22-A-D	de novo	TOP
4	29	17+ 5	18.15	del(22q11.21,1.39 M)	18,778,537–20,164,052	arr[GRCh37] 22q11.21(18,648,856_21,464,764)x1	2816 Kb	LCR22-A-D	de novo	TOP
5	32	18+ 1	25.23	del(22q11.21,2.03 M)	18,476,201–20,501,581	arr[GRCh37] 22q11.21(18,648,856_20,312,661)x1	1664 Kb	LCR22-A-B	de novo	TOP
6	30	14+ 6	10.48	del(22q11.21,1.72 M)	18,778,537–20,501,581	arr[GRCh37] 22q11.21(18894835_20311763)x1	1417 Kb	LCR22-A-B	de novo	TOP
7	34	20	21.67	del(22q11.21,1.72 M)	18,778,537–20,501,581	arr[GRCh37] 22q11.21(18,916,843_20,312,661)x1	1396 Kb	LCR22-A-B	de novo	TOP
8	34	16+ 2	20.03	del(22q11.21,1.72 M)	18,778,537–20,501,581	arr[GRCh37] 22q11.21(18,970,562_21,058,888)x1	2088 Kb	LCR22-A-D	de novo	TOP
9	29	16+ 5	13.79	del(22q11.21,2.03 M)	18,476,201–20,501,581	Normal	–	–	–	Live birth
10	35	16+ 6	17.31	del(22q11.21,1.54 M)	18,957,712–20,501,581	Normal	–	–	–	Live birth
11	20	19+ 1	12.78	del(22q11.21,1.04 M)	19,462,422–20,501,581	Normal	–	–	–	Live birth
12	30	17	20.51	del(22q11.21,1.04 M)	19,462,422–20,501,581	Normal	–	–	–	Live birth
13	20	17	14.25	del(22q11.21,3.57 M)-M	18,476,201 22,050,974	Normal	–	–	–	Live birth
14	25	16+ 6	12.70	del(22q11.21,2.89 M)-M	18,778,537–21,663,666	Normal	–	–	–	Live birth
15	32	16+ 2	7.49	del(22q11.21,1.44 M)	19,061,538–20,501,581	Normal	–	–	–	Live birth
16	26	16	15.58	del(22q11.21,2.89 M)-M	18,778,537–21,663,666	Normal	–	–	–	Live birth
17	24	16+ 4	12.57	del(22q11.21,1.59 M)	18,476,201–20,063,469	arr[GRCh37] 4q21.1(76,743,797_78,616,609)x3	1873 Kb	–	–	Live birth
18	27	15+ 4	15.51	del(22q11.21,2.89 M)	18,778,537–21,663,666	Refuse invasive testing	–	–	–	CHD
19	30	13+ 5	29.22	del(22q11.21,1.04 M)	19,462,422–20,501,581	Refuse invasive testing	–	–	–	TOP
20	28	14+ 1	16.23	del(22q11.21,1.14 M)	19,362,403–20,501,581	Refuse invasive testing	–	–	–	TOP
21	41	13	17.02	del(22q11.21,1.1 M)	19,161,822–20,265,139	Refuse invasive testing	–	–	–	TOP
22	37	13+ 6	13.40	del(22q11.21,1.39 M)	18,578,067–19,963,378	Refuse invasive testing	–	–	–	Live birth

GA, gestigate age; wk, weeks; cffDNA, cell-free fetal DNA; -M, Maternal derived; TOP, terminate of pregnancy; CHD, congenital heart disease.

In addition, this study identified four cases of 22q11.2 duplication, with an incidence of 0.010%. All duplications were larger than 3 Mb and were suspected to originate from maternal CNVs (Supplementary Table 2).

Amniocentesis for CMA analysis

Of the 22 high-risk cases for 22q11.2 deletion identified by NIPS, 17 (17/22, 77.27%) underwent amniocentesis. CMA confirmed 22q11.2 deletion in 8 cases, giving a positive predictive value (PPV) of 47.06% (8/17) for the detection of 22q11.2 DS by NIPS (Table 2; Fig. 3). Nine fetuses were not identified: one showed a 1873 Kb duplicated CNV in the 4q21.1 region, and the other eight cases had no pathogenic CNVs detected by CMA (Table 2). Of note, a maternal 22q11.2 deletion was confirmed in three fetus mothers. In addition, two of the four high-risk cases for 22q11.2 duplication were confirmed positive by CMA, resulting in a PPV of 50%(2/4) (Supplementary Table 2).

Fig. 3.

CMA results. CMA results for 8 high-risk cases with 22q11.2 deletion screening are displayed. Cases 5, 6, 7, and 8 exhibit smaller deletion fragments than cases 1–4. Cases 27 and 28 are screen-negative for 22q11.2DS but have ultrasound abnormalities; CMA confirms 22q11.2DS in these cases (black arrows). Cases 25 and 26 are at high risk for 22q11.2 duplication with positive CMA results (black arrows). Green rectangles denote disease-associated genes; red rectangles indicate 22q11.2 deletions; blue rectangles indicate 22q11.2 duplications.

Genetic origin and CMA results

CMA confirmed 22q11.2 deletion in eight high-risk cases identified by NIPS, all of which were de nove variants confirmed by multiplex ligation-dependent probe amplification (MLPA). Among these eight true positives, CMA detected five cases with an LCR22A-D deletion, and three with an LCR22A-B deletion. Of the two cases with 22q11.2 duplications, one had an LCR22A-D duplication, and one had an LCR22 F-H duplication (Table 2; Fig. 3).

In cases where CMA was negative, NIPS suggested the presence of maternal CNV, and MLPA confirmed 22q11.2 deletion in the mothers, but these variants were not transmitted to the fetuses. In addition, four cases with NIPS suggesting a maternal 22q11.2 duplication were confirmed by MLPA in two pregnant women (Supplementary Fig. 3). As fluorescence in situ hybridization (FISH) was not performed to detect mosaicism, parents who tested negative by MLPA can not rule out the possibility of being carriers of the 22q11.2 deletion or duplication.

Karyotyping analysis

Karyotype analysis was normal in all fetuses examined.

Pregnancy outcomes and follow-up

Of the 22 high-risk 22q11.2 deletions, eight pregnancies were terminated after genetic counseling. Three of these cases had no ultrasound abnormalities, while five had fetal cardiovascular abnormalities.

Of the nine false-positive cases, one had a copy number duplication at 4q21.1 and was delivered by cesarean section at 39^+ 4 weeks, weighing 3410 g with an Apgar score of 10 (Table 2). The remaining eight cases, with normal CMA results, continued their pregnancies and delivered healthy infants (Table 2).

Five cases declined prenatal diagnosis: one case was terminated due to ultrasound abnormalities, while two other cases opted to terminate despite normal ultrasound findings. The remaining two cases continued their pregnancies, resulting in one fetus with congenital heart disease (ventricular septal defect and atrial septal defect) (Supplementary Fig. 4).

Follow-up of low-risk cases

Among the 38,081 low-risk cases, two cases of 22q11.2 deletion and two cases of 22q11.2 duplication were confirmed by CMA due to abnormal ultrasound findings. The deletions and duplications measured 3164 Kb, 1970 Kb, 2811 Kb, and 1385 Kb, respectively (Supplementary Fig. 2, Supplementary Table 2), with terminations occurring in all these cases. The cffDNA concentrations were 6.99% and 8.23% for deletions and 19.87% and 23.23% for duplications (Supplementary Table 3).

Discussion

22q11.2 deletion syndrome (22q11.2DS) is the most common microdeletion syndrome, with an incidence of approximately 1/800–1/1000 in fetuses and 1/3000–1/6000 in live births^4,6,18,22. The syndrome has a wide variety of clinical features, which vary considerably even within the same family. Major clinical manifestations include congenital heart defects(64%), palatal anomalies (67%), immunodeficiency (77%), developmental delay and learning disabilities (70–90%), psychiatric disorders such as autism spectrum disorder(20%), schizophrenia (25%), and other rare phenotypes, such as subglottic stenosis, imperforate anus, esophageal atresia^23–25. This variability complicates diagnosis and management, and highlights the importance of early detection.

Prenatal diagnosis of 22q11.2 DS typically relies on the identifying cardiovascular abnormalities on fetal ultrasound^24,26,27. Other ultrasound findings may include thymic hypoplasia, craniofacial anomalies, renal anomalies, increased nuchal translucency, and polyhydramnios^27–29. However, detailed ultrasound scans are usually performed around 20–24 weeks of gestation, which may limit early detection. In this study, NIPS detected eight cases of 22q11.2 deletion between 14 and 19 weeks of gestation; only three cases showed cardiovascular abnormalities on subsequent ultrasound. This finding highlights the utility of NIPS in early detection, which can be critical as reliance on ultrasound alone may lead to missed diagnoses or delayed detection.

The prevalence of pathogenic CNVs in the general pregnant population can be as high as 1.6–1.7%, even in the presence of normal ultrasound findings, and is not strongly correlated with CNVs or maternal age^30–32. NIPS has become an essential tool in prenatal genetic testing, particularly for the detection of chromosomal abnormalities such as trisomies and microdeletions. The ACMG has recommended its use for screening for 22q11.2 deletion syndrome (22q11.2 DS) in all pregnant women. However, further studies are needed to validate the effectiveness of NIPS for 22q11.2 deletion in the general pregnant population in China. In regions such as Shenzhen City in Guangdong Province, NIPS is offered to all pregnant women as part of a public health initiative, often at minimal or no cost. Consequently, the study cohort comprises an unselected population, providing a broader perspective on the performance of NIPS for 22q11.2 deletion in a diverse, general population.

Different technologies have been used for NIPS to detect 22q11.2 deletions, offering different levels of sensitivity and positive predictive value (PPV). For example, an SNP-based method achieved a PPV of 52.6% and a false positive rate of 0.05%¹⁸, semiconductor sequencing showed a PPV of 75%¹⁵, and Illumina’s NIPS-PLUS achieved a PPV of 97%¹⁶. These variations highlight the advances and differences in NIPS technology. Compared to other methods, our study using the DNBSEQ platform showed a sensitivity of approximately 83.33% (10 out of 12 cases) and a PPV of 47.06% (8 out of 17 cases). This is lower than reported for semiconductor sequencing platforms and Illumina’s NIPS-PLUS^15,16. While the DNBSEQ platform proved effective for routine screening in a general population, it showed relatively lower sensitivity and PPV than these more advanced platforms. However, the DNBSEQ method offers advantages in terms of cost-effectiveness and accessibility, making it suitable for large-scale population screening in resource-limited settings.

This study screened 22 high-risk 22q11.2 deletion cases from a general pregnant population of 38,495 women. Each sample had an average of approximately 8.34 million unique sequencing reads and an average fetal cell-free DNA concentration of 16.42%. 17 cases were confirmed by prenatal diagnosis, with 8 true positives and 9 false positives. Two out of the 5 high-risk cases did not undergo prenatal diagnosis, and subsequent follow-up revealed significant cardiovascular ultrasound abnormalities. The PPV of 47.06% indicates that almost half of the high-risk results could be false positives, highlighting the importance of confirmatory prenatal diagnostic techniques such as chromosomal microarray analysis (CMA) or multiplex ligation-dependent probe amplification (MLPA).

Interestingly, although NIPS successfully detected most 22q11.2 deletions, a few of cases were missed, particularly in low-risk pregnancies. Postnatal diagnosis identified two false-negative cases, suggesting that routine screening may miss some deletions in populations with lower risk factors. These results are consistent with other studies that have reported a range of sensitivities from 75.0% to 99.8% for NIPS screening for 22q11.2 deletion^{4,11,18,33–35}. The factors leading to false-negative results in NIPT screening mainly include low concentration of fetal free DNA, restricted placental chimerism, low sequencing depth or the influence of bioinformatics algorithms, as well as the interference from maternal copy number variations. In addition, Case 17 which was identified as high-risk for 22q11.2 deletion by NIPS but confirmed to have a 1873 Kb duplication in the 4q21.1 region via CMA, raises concerns about genomic background interference in NIPS, and low-depth sequencing (average 8.34 million unique reads in this study) may lead to misinterpretation of signals.

A notable limitation of our study is the short-term follow-up of 22q11.2 deletion cases classified as low risk. We observed two false-negative cases, which may indicate that some 22q11.2 deletion cases remain undetected in low-risk populations. The sensitivity of routine screening may be lower than the observed 83.33%. In addition, cases with low cffDNA concentrations (< 10%) could contribute to false negatives, highlighting the need for improved enrichment techniques. For the 22q11.2 duplication, the challenge is even greater. Screening may fail to detect 22q11.2 duplication cases even when cffDNA concentrations reach 20%, making clinical application of routine screening for this condition difficult.

Another limitation is the incomplete collection of clinical information. Due to the retrospective design, detailed ultrasound findings (e.g., fetal conotruncal malformations, thymic hypoplasia) and genetic counseling records were not systematically collected for all cases, which hinders the integration of laboratory data with clinical management. Future prospective studies should establish an Multidisciplinary Team (MDT) system to collect full-process clinical data and develop standardized clinical pathways for 22q11.2 CNV screening and diagnosis.

Our findings emphasize the importance of genetic counseling and further prenatal diagnosis testing following a high-risk NIPS result. The PPV of 47.06% in our study indicates that almost half of the high-risk results are false positives, so it is crucial to provide pregnant women with accurate information about the clinical implications of the results and the availability of confirmatory testing. Genetic counseling is particularly important for those whose NIPS results suggest maternal CNVs, as these findings may not affect the fetus but require further maternal genetic investigation. In addition, NIPS only represents placental cells and does not test for fetal-derived DNA, so many problems such as placental mosaicism, maternal SCA, maternal cancer, vanishing twins, and maternal obesity will never be resolved by NIPS.

In conclusion, our study confirms that routine NIPS for 22q11.2 deletion has significant clinical value in early detection with a PPV of 47.06% and a sensitivity of 83.33%. Despite its limitations, including false negatives and the need for further genetic counseling and diagnosis, routine NIPS remains an effective tool for large-scale population screening, especially in areas where access to more advanced technologies is limited. Further refinement of NIPS technologies and integration of maternal DNA analysis may improve detection rates and reduce false positives, ultimately increasing its clinical utility.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

Conceptualization, data analysis, and manuscript writing: X.Y.C., L.H., Y.Y.P., and W.Q.L. Clinical data collection and experiment: J.T.Z., L.J.W., Y.N.L., T.Z.; Ultrasound data collection: J.S.S.; Drafting: Y.Y.P., and W.Q.L.; Writing review and editing: W.Q.L. All authors read and approved the manuscript and are accountable for all study aspects.

Funding

This research was funded by the Medical and Health Technology Research Project of Longgang District, Shenzhen City (grant LGWJ2023-56), the Medical and Health Technology Research Project, Special Funds for Science and Technology Innovation in Longgang District, Shenzhen City (grant LGKCYLWS2022013), and the Natural Science Foundation of Shenzhen City (JCYJ20230807141908017).

Data availability

Data are available upon reasonable request from the corresponding author.

Declarations

Competing interests

The authors declare no competing interests.