Chromosomal copy number analysis of products of conception by conventional karyotyping and next‐generation sequencing

Abstract Purpose Chromosomal abnormalities are a major cause of spontaneous abortion, and conventional G‐banded karyotyping (G‐banding) is mainly utilized for chromosomal analysis. Recently, next‐generation sequencing (NGS) has been introduced for chromosomal analysis. Here, we aimed to investigate the applicability and utility of NGS‐based chromosomal analysis of products of conception (POC) on chorionic villus samples from spontaneous abortion. Methods The results of chromosomal analysis of 7 chorionic villus samples from spontaneous abortion were compared between conventional G‐banding and NGS‐based chromosomal copy number analysis. Age dependency and frequency of each chromosomal aneuploidy were evaluated for 279 cases analyzed by NGS. Results Excluding two cases (culture failure and maternal cell contamination), the results were consistent between G‐banding and NGS. For cases analyzed by NGS, the rate of chromosomal abnormality increased in a maternal age‐dependent manner. The frequency of each chromosomal aneuploidy detected by NGS was almost the same as that previously reported. Finally, NGS analysis was possible for difficult cases by G‐banding analysis, such as culture failure, maternal cell contamination, long‐term storage cases, and low cell number. Conclusions Chromosome analysis using NGS not only obtains comparable results to conventional G‐banding, but also can analyze POC more accurately and efficiently.


| INTRODUC TI ON
The rate of spontaneous abortion is 10%-15% in clinically recognized pregnancies, and chromosomal abnormalities, such as aneuploidy, polyploidy, structural rearrangement, duplication, or deletion, are the most common reason. [1][2][3] Conventional cytogenetic analysis by G-banded karyotyping (G-banding) is generally performed for chromosomal analysis, but it has several disadvantages such as low resolution, the necessity of long-term cell culture, and enough experience for diagnosis. Moreover, cell culture failure or maternal cell contamination often occurs, which are significant obstacles for the analysis. [3][4][5] Currently, molecular cytogenetic approaches, such as array comparative genome hybridization and next-generation sequencing (NGS), have been introduced for comprehensive chromosomal analysis. 3,4,6 Among them, NGS has attracted attention as an accurate, high-resolution, and high-throughput method and is expected to be applicable for chromosomal analysis of products of conception (POC) from spontaneous abortion. [7][8][9][10] In Japan, it is generally known that elderly pregnancy and late birth have been increased in association with several social factors, such as advancing of late marriage. Number of miscarriages in spontaneous pregnancy is speculated to be increased owing to the chromosomal abnormalities with advanced maternal age. In order to deal with the increasing number of chromosomal analysis of POC, more accurate and high-throughput test methods are required such as NGS. However, to our knowledge, no reports have verified the utility of NGS-based chromosomal analysis of POC in Japan. Therefore, in this study, we aimed to compare the consistency between conventional G-banding and NGS-based chromosomal copy number analysis of POC using chorionic villus samples. In addition, age dependency and frequency of each chromosomal aneuploidy were evaluated for 279 cases analyzed by NGS. Finally, points of attention for NGS-based chromosomal analysis were discussed.

| Sample collection
Two studies were performed in this report, one was comparison of the consistency between conventional G-banding and NGS-based chromosomal copy number analysis of POC. From February 2018 to April 2018, chromosomal analysis of 7 chorionic villus samples from spontaneous abortions (7-9 weeks) was carried out using both conventional G-banding and NGS. Another was evaluation of age dependency and frequency of each chromosomal aneuploidy for 279 cases from February 2018 to December 2019 analyzed by NGS.

| G-banding and NGS
Conventional G-banding analysis was carried out by Nihon Gene Research Laboratories Inc. For NGS-based chromosomal copy number analysis, chorionic villi were collected from POC. To prevent maternal cell contamination, only distinct chorionic villus was collected under the stereomicroscope and maternal tissue (eg, endometrium or peripheral blood) was removed as much as possible.
Genomic DNA was isolated using the QIAamp DNA Mini Kit (Qiagen GmbH) according to the manufacturer's instructions. Library construction was performed using the VeriSeq PGS kit (Illumina), and the MiSeq system (Illumina) was used for DNA sequencing following the manufacturer's protocol with minor modification. G-banding and NGS analysis were carried out in different laboratories which were registered clinical laboratory by Japan Registered Clinical Laboratories Association. Each laboratory performed quality control internally and externally; however, G-banding and NGS analysis were different not only quality control method but also detection principle and obtained results. Thus, we had to comprehensively consider the data quality, specificity, possibility of contamination, and so on, when obtained inconsistent test results.

| Data analysis
Each chromosomal copy number was comprehensively estimated using the Bluefuse Multi Analysis Software (Illumina). This platform was demonstrated to have a detection performance of around 10 Mb of segmental aneuploidy and 20% of mosaicism. The range of mosaicism was defined as 20%-80%, which complied with Preimplantation Genetic Diagnosis International Society guidelines. 11,12 Statistical analysis was performed using Fisher's exact test with the Holm correction for multiple testing, and differences were considered significant at P < .05.

| RE SULTS
Comparison of results obtained from conventional G-banding and NGS is summarized in Table 1. NGS successfully analyzed all seven cases.
Conversely, G-banding only detected six cases; one case was undetectable because of cell growth failure. In the six cases analyzed by G-banding, the results of five cases were consistent with the results of NGS. However, one case (Case 7) was suspected to have maternal cell contamination (G-banding: 46,XX, NGS: 46,XY). Among the seven cases analyzed by the NGS, two cases (Cases 4 and 7) had normal male karyotype (46,XY) and five cases exhibited autosomal trisomy, implying that there were no cases suspected of maternal cell contamination.
Finally, the frequency of each chromosomal aneuploidy in the 186 abnormal cases was investigated, as shown in Figure 2.

| D ISCUSS I ON
Over 50% of miscarriages are the result of chromosomal abnormalities. [1][2][3][4] As described in Table 1, two cases of 46,XY were recurrent pregnancy loss cases (Cases 4 and 7), demonstrating the possibility that these abortions did not result from chromosomal abnormalities but from another reason(s) for recurrent miscarriage. The other five cases were autosomal trisomies, indicating that chromosomal abnormalities were the most common reason for spontaneous abortion.
These findings are consistent with those of previous reports. [1][2][3][4] It has been reported that culture failure in G-banding occurs not only due to technical problems such as minimal chorionic villi and/or prolonged time from diagnosis of miscarriage to dilation and curettage, but also because of growth arrest derived from chromosomal abnormality itself. 13 One case of culture failure in this study was blighted ovum (Case 1); thus, NGS might be useful for difficult cases of cell culture such as specimens with few chorionic villi. In addition, NGS could analyze POC stored a long time after miscarriage, for instance, samples stored in saline at 4°C for a month (Case 7). NGSbased analysis not only obtained equivalent results to G-banding, but also was able to analyze specimens that were difficult to detect by G-banding. However, additional comparative evaluations are needed due to the small number of cases included in this study.
Although villus-like fibrous tissues could be grossly seen, sometimes tissues other than chorionic villus were observed under the stereomicroscope, as shown in Figure S1. The NGS result of these fibrous tissues indicated normal female (46,XX); therefore, maternal cell contamination was strongly suspected. Histopathologic examination suggested that most fibrous tissues were large vessels or endometrium with morphological changes caused by the action of progesterone (data not shown). In the case of blighted ovum, small or unclear fetus, and miscarriage at home, distinct chorionic villi were often not identified. Even in such cases, it is important to collect distinct chorionic villi for accurate testing, and small villi would occasionally be found in the saline used for rinsing ( Figure S2). These observations suggested that the accuracy of the analysis might be improved by using tissue fragments and rinsing solution in specimens without distinct chorionic villi.
The rate of maternal cell contamination on conventional G-banding has been reported. 5 Lathi et al 14

ACK N OWLED G M ENTS
This study was supported by a Grant-in-Aid for Encouragement of Scientists of the Japan Society for the Promotion of Science (JSPS) (Grant Number 19H00459).

CO N FLI C T O F I NTE R E S T
The authors declare no conflicts of interest associated with this research.

H U M A N R I G HT S A N D I N FO R M E D CO N S E NT
All procedures followed were in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national) and with guidelines set forth by the Helsinki Declaration of 1964 and its later amendments. The study was approved by the clinical ethics committee, and informed consent was obtained from all patients for inclusion in this study.

A N I M A L S TU D I E S
This study does not contain any studies with animal subjects performed by any of the authors.