Combined diagnosis of QF‐PCR and CNV‐Seq in fetal chromosomal abnormalities: A new perspective on prenatal diagnosis

Abstract Objective This study aimed to evaluate the effect of QF‐PCR and CNV‐seq in diagnosing prenatal fetal chromosomal aberrations, explore the advantages and necessity of multimethod joint diagnosis. Methods We chose pregnant women with the indication of fetal chromosome examination in our hospital last year, collected 657 cases of amniotic fluid for QF‐PCR and CNV‐seq analyzes. Results While detecting aneuploidy, the coincidence rate of QF‐PCR and CNV‐seq was 100% (56/56). For all 46 chromosomes, 523 cases (79.60%, 523/657) coincided precisely, 128 cases (19.48%, 128/657) showed abnormality with CNV‐seq, 8 cases (1.22%, 8/657) revealed abnormality by QF‐PCR. In serological Down's syndrome screening, 328 cases showed a high risk of trisomy 21, of which CNV‐seq and QF‐PCR were consistent in 4 cases (1.22%, 4/328), CNV‐seq found 87 cases of CNVs in 78 samples except for chromosomal aneuploidy abnormalities, among these, 18 cases (20.69%, 18/87) were polymorphic, 7 cases (8.05%, 7/87) might cause disease, 13 cases (14.94%, 13/87) caused disease explicitly, 21 cases (24.14%, 21/87) were possibly benign, 17 cases (19.54%, 17/87) were explicitly benign, and the classification of 11 cases (12.64%, 11/87) was unclear. Conclusion QF‐PCR and CNV‐seq were highly consistent in diagnosing chromosomal aneuploidy. The high risk of serological Down's screening might not only due to the aneuploidy of chromosomes 21, 18, and NTD, but also the microdeletion or microduplication of all 46 chromosomes. So using CNV‐seq combined with QF‐PCR could effectively reduce the risk of missed diagnosis.


| INTRODUC TI ON
Birth defects, also known as congenital abnormalities, are the leading cause of neonatal morbidity and death globally. 1 Such defects refer to abnormalities in the individual, morphology, structure, and function (including metabolism, mentality, and intelligence) that occur in the uterus before the birth of the fetus but are not caused due to childbirth injuries. Such defects include congenital malformations, genetic metabolic defects, congenital disabilities (blindness, deafness, and dumbness), immune diseases, mental retardation, etc. Overall, birth defects affect approximately 1 in 33 children. 2 According to reports, primary prevention is not possible as the cause of around 60% of birth defects is unknown, and about 20% of congenital disease are due to genetic defects. 3 The most common genetic disease that causes birth defects in newborns is chromosomal abnormalities, accounting for about 1/160 live birth. 4 Chromosome abnormalities, in general, include an abnormal number of chromosomes and abnormal chromosome structure. 21-trisomy syndrome (Down syndrome), 18-trisomy syndrome (Edward syndrome), 13-trisomy syndrome (Patau syndrome), and sex chromosome aneuploidy (SCA) are due to a common abnormal number of chromosomes. 5 The deletion or duplication of a tiny fragment of chromosomes is a common chromosomal structural abnormality, which is the main reason of birth defects in newborns. Nearly 300 such diseases have been found until now, such as DiGeorge, Prader-Willi, Angelman, and Williams syndrome, Williams-Beuren syndrome, 17q21.31 microdeletion syndrome, Prader-Willi and Angelman syndrome, etc. 6,7 At present, there is no effective treatment for chromosomal abnormalities.
However, more and more technologies have been applied for the early diagnosis of chromosomal abnormalities, in order to achieve early prenatal intervention.
The gold standard for diagnosing fetal chromosomal abnormalities is the karyotype till now, which analyzes cells extracted from the amniotic fluid. Fetal chromosomal aneuploidy, polyploidy, abnormal balance structure, chimera, and deletions and duplications that are bigger than 10-20 Mb can be diagnosed through it. However, cell culture is required, and the method has many shortcomings such as long detection time, low throughput, and the inability to detect copy number variations (CNVs) below 5 Mb. 8,9 In recent years, there has been wide usage of genome copy number variation sequencing (CNV-seq) technology based on lowdepth whole-genome sequencing due to its high throughput, simple operation, and only a small sample required for the detection of chromosomal aberrations, including aneuploidy, microdeletion, microduplication, etc. Based on short tandem repeat (STR), the quantitative fluorescent polymerase chain reaction (QF-PCR) technology has high throughput, high speed, and high accuracy and the ability to detect the contamination of maternal blood.
In this study, prenatal screening (serological Down's screening, noninvasive DNA testing, ultrasonography, etc.) and prenatal diagnosis (CNV-seq, QF-PCR, chromosome karyotype analysis, etc.) were carried out on 657 pregnant women with indications for chromosome examination. Then conducted a comparative analysis to explore the respective advantages of the above-mentioned techniques and the possible clinical significance of the abnormal results.

| Materials
For this study, we selected pregnant women with singleton who came to our hospital for prenatal consultation from July 30, 2019 to October 23, 2020 due to abnormal serological Down's syndrome screening, high risk of noninvasive prenatal testing (NIPT), abnormal B-ultrasound, family genetic history, adverse pregnancy history, and other factors (taking medication during pregnancy, in vitro fertilization-embryo transform, etc.). These women signed the informed consent. We then collected 657 amniotic fluid samples and performed CNV-seq and QF-PCR tests at the same time. Karyotype analysis were conducted of some specimens.

| QF•PCR analysis
We took 1.9 ml of amniotic fluid and centrifuged it to remove the supernatant, then used a magnetic bead method nucleic acid extraction kit (Guangzhou Darui Biotechnology Co., Ltd.) to extract the genomic DNA from the amniotic fluid and a K5800 microspectrophotometer to detect DNA quality and concentration. Then, we stored the sample at −20℃. On referring to the instructions of the corresponding kit, we detected and analyzed the samples. After amplification, we took 1 μl of the PCR product and mixed 13.5 μl formamide and 0.5 μl Liz600. Then analyzed the fragment by ABI3500DX, GeneMapper5.0, for data analysis.

| CNV•seq analysis
The Hunan Jiahui Genetics Specialist Hospital did CNV-seq analysis of amniotic fluid. The experimental steps were as follows. DNA was extracted from the amniotic fluid and hydrolyzed with restriction enzymes to obtain DNA fragments with an average size of 200 bp.
A library was prepared using the PCR-free method (Beijing Beiruihe Kang Biotechnology Co., Ltd.) and connecting adapters. Then, 36 bp single-end sequencing was done on a high-throughput sequencing platform (NextSeq CN500 platform, Illumina), with a depth of 0.1×. All determined sequences were aligned and analyzed with the hg19 human genome through parallel alignment software (using the Burrows and Wheeler algorithm). 10 Using 100 kb as the basic unit of analysis, the human genome was divided into several continuous regions, and the number of unique reads that matched in each region was counted. To determine the CNVs of the sample, a unique algorithm was used. According to the statistical results, the normalized sequence copy number was on the y-axis. The continuous 100 kb Duplications were defined as copy number (CN) >2. 8

| The classification of etiology
We classified the etiology of patients who underwent amniocentesis. Figure 1 Table 1. Therefore, the overall coincidence rate of QF-PCR and CNV-seq in the diagnosis of 21, 18, 13, and sex chromosome aneuploidy was 100%.

| Comparison of QF-PCR and CNV-seq in the diagnosis of all chromosomal aberrations
Since QF-PCR can only detect 21, 18, 13, sex chromosome aneuploidy, while CNV-seq can detect all chromosomes, the coverage of QF-PCR is far less than that of CNV-seq. In the diagnosis of all chromosomal aberrations, 523 cases (79.60%, 523/657) showed consistent in CNV-seq and QF-PCR. CNV-seq showed 128 cases (19.48%, 128/657) as abnormal, but QF-PCR could not detect these. QF-PCR showed 8 cases (1.22%, 8/657) as abnormal, but CNV-seq could not detect these. The corresponding karyotypes were listed in Table 2.

| CNVs that could not be detected by QF-PCR but were detected by CNV-seq
The QF-PCR kit used in this study can only detect aneuploidies of the five chromosomes (21, 18, 13, X, and Y chromosomes). There were 141 cases CNVs existed in 128 samples that QF-PCR could not detect but were detected by CNV-seq, other than five chromosomal aneuploidies (among them, more than one kind of abnormal F I G U R E 1 The classification of etiology The classification of 29 cases (20.57%, 29/141) was not clear, out of which 5 cases were located on 5 chromosomes, and 24 cases were located on other chromosomes. The corresponding karyotypes were listed in Table 3.
Among the 20 cases of pathogenic CNVs in 19 samples, 4 cases of CNVs had microdeletions or microduplications, which were >5 Mb, but the remaining 16 cases were <5 Mb, below the detection limit of karyotype analysis (5 Mb); therefore, it is impossible to detect microdeletions and microduplications smaller than 5 Mb using karyotype analysis, like the 16 cases pathogenic CNVs detected by CNV-seq in this study. Table 4 included the results of CNV-seq and pathogenic information.

| Abnormalities that could not be detected by CNV-seq but were detected by QF-PCR
We found 8 samples containing 9 cases of microduplications that could not be detected by CNV-seq but were detected by QF-PCR.
The other effective STR sites on the chromosomes were all normal. However, the clinical significance was still unclear. As shown in Table 5, it needs to be diagnosed jointly with clinical and other tests.   One or more CNVs might be present in the same sample, so the number of CNVs was greater than the number of samples.

TA B L E 3 Pathogenicity analysis of CNVs that could not be detected by QF-PCR but detected by CNV-seq
were possibly benign, and the classification of 5 cases (25.00%, 5/20) was still unclear.

| DISCUSS ION
Each year, there have been roughly 135 million newborns worldwide, of which 3% suffer from major structural birth defects. This brings a serious economic and spiritual burden to society and the family. 13 Therefore, diagnosing fetal chromosomal aberrations quickly and accurately has become more and more important to eliminate the mother's anxiety and reduce the birth rate of abnormal fetuses.
Till now, karyotype analysis is the gold standard for prenatal diagnosis of fetal chromosomal aberrations. However, there are many shortcomings, such as long detection time, cell culture requirement, low resolution (<5 Mb), and misdiagnosis due to maternal blood contamination. 14 CNV-seq technology has been widely used in detecting chromosomal aberrations such as aneuploidy, microdeletion, and microduplication due to its many advantages, such as high throughput and easy operation. However, it cannot identify maternal cell Abbreviations: CNV-seq, copy number variation sequencing; QF-PCR, quantitative fluorescent polymerase chain reaction.
a Two pathogenic CNVs were present in the same sample.
contamination. QF-PCR can diagnose common chromosomal aneuploidy within 24 h, through qualitative and quantitative analysis of the polymorphism of STR genetic markers, adopt multiplex PCR amplification and capillary electrophoresis separation technology.
One of the advantages of QF-PCR is that it can identify maternal contamination. However, it cannot detect chromosome structure abnormalities, chromosome polyploidy, and mosaics with a mosaic ratio that is <20%. 15 Among the 657 amniotic fluid specimens in this study, while diagnosing aneuploidy in 5 chromosomes (13, 18, 21, X, and Y), we found 33 cases of trisomy 21, 5 cases of trisomy 18, 1 case of trisomy 13, 4 cases of XXX, 7 cases of XXY, and 6 cases of XYY. The coincidence rate between CNV-seq and QF-PCR was 100%. For all chromosomal structure and number abnormalities, the rate of QF-PCR and CNV-seq was the same at 79.60%. The rate of abnormality indicated by CNV-seq but not QF-PCR was 19.48%, whereas that indicated by QF-PCR but not CNV-seq was 1.22%.
Among the 9 cases abnormalities that could not be detected by  One or more CNVs might be present in the same sample, so the total number was greater than the number of samples. rate of chromosome 18 was 33.33% (3/9), the coincidence rate of chromosome 13 was 20.00% (1/5), and the coincidence rate of sex chromosomes was 39.39% (13/33). The accuracy rate was improved compared to serological Down's syndrome screening, but it is still a screening experiment that could not be made for a final diagnosis.
In addition, the accuracy of NIPT decreased correspondingly in the following situations: if the pregnancy week was too early or too late, if the expected age was ≥35 years, if the pregnant woman was severely obese (body mass index >40), etc. 18 In conclusion, QF-PCR can detect the contamination of mater-

| Limitation statement about the research
This research based on a small sample size, so the conclusion might have limited generalizability, we will continue to collect samples in future to get more convincing results. In addition, our research is only for Chinese people, and we encourage scientists from other countries to also participate in this research.

ACK N OWLED G M ENT
We thank the patients who participated in this study.

AUTH O R CO NTR I B UTI O N S
All authors have accepted responsibility for the entire content of this manuscript and approved its submission.

CO M PE TI N G I NTER E S TS
Authors state no conflict of interest.

I N FO R M ED CO N S ENT
Not applicable.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author (dyq13655602140@163.com) upon reasonable request.