Consistency between chromosomal status analysis of biopsied human blastocyst trophectoderm cells and whole blastocyst cells

Abstract Purpose This study investigated the consistency between results of preimplantation genetic testing for aneuploidy performed on trophectoderm (TE) cells and remaining blastocyst cells. Methods TE biopsy was performed on 29 surplus cryopreserved human blastocysts. Biopsy samples and remaining blastocysts were processed using the VeriSeq PGS kit, and chromosomal statuses were compared by next‐generation sequencing. Results Discordance was observed in the chromosomal status of 11 out of 29 blastocysts between the biopsied TE and remaining blastocysts. Concordance was observed in 11 of 12 blastocysts classified as euploid by TE biopsy and in 7 of 17 blastocysts classified as aneuploid. There was 100% concordance (7/7) in cases diagnosed as aneuploid with no mosaicism by TE biopsy. However, discordance was observed in all 10 cases showing mosaicism or partial chromosomal abnormality. Conclusion Chromosomal status analysis based on TE biopsy does not accurately reflect the chromosomal status of the whole blastocyst. The chromosomal status is usually the same between the TE and remaining blastocyst cells in cases diagnosed as euploid or aneuploid with no mosaicism. However, mosaic blastocysts and those with other types of structural rearrangements have a higher risk of inconsistency, warranting caution during embryo selection.


| INTRODUC TI ON
In assisted reproductive technology (ART), preimplantation genetic testing for aneuploidy (PGT-A) of blastocysts is widely used to improve the rate of implantation and lower the rate of miscarriage by selecting euploid embryos for in vitro fertilization and embryo transfer therapy. [1][2][3][4][5] Guzman et al reported that no more than 5 trophectoderm (TE) cells should be sampled for PGT-A, 6 as sampling numerous cells might damage blastocysts and increase the risk of implantation failure and miscarriage. 6,7 They also found that approximately 5 biopsied TE cells are sufficient for performing chromosomal status analysis.
Human blastocysts are constituted by approximately 300 cells; 8 however, to this day, no study has determined the extent to which the 5 biopsied TE cells represent the chromosomal status of the whole blastocyst. As is known, TE cells differentiate into placental tissues after implantation. Therefore, knowledge of the chromosomal status of the inner cell mass (ICM), which significantly contributes to embryonic differentiation, is important. The chromosomal status of TE and ICM has been analyzed separately and compared in several studies. [9][10][11] Nonetheless, these reports were based on partial biopsy of the ICM and TE and did not consider any specific associations between the outcomes of the TE biopsy and ICM chromosomal status or whole-embryo mosaic rates. Furthermore, none of these studies managed to elucidate a clear relationship between the chromosomal status of TE biopsy samples and the group of ICM cells that will later form the fetus. 12 Of note, invasive procedures for determining the chromosomal Despite recent reports on PGT-A using cell-free DNA (cfDNA) from spent embryo culture medium, [14][15][16] the consistency between the actual blastocyst chromosomal status and the origins of cfDNA has been debatable. At present, PGT-A implemented globally in clinical practice uses TE biopsy samples. Therefore, in this study, we used next-generation sequencing (NGS) to analyze the relationship between the chromosomal status profiles of approximately 5 biopsied TE cells and the remaining blastocyst cells from 29 frozen and thawed human blastocysts. In particular, we compared the chromosomal data obtained from biopsied TE cells and the remaining blastocyst cells to determine the consistency and association between the 2 sets of chromosomal data.

| Embryo source and study participant details
Twenty-nine blastocysts obtained from 11 women were used in this study. Blastocysts were prepared and frozen as part of an ART according to the Gardner classification. 17 The mean age of patients who provided their blastocysts for freezing was 34.7 ± 2.7 years.
Patients had been infertile for 4.0 ± 2.4 years, and the average number of retrieved oocytes was 2.4 ± 1.8. In all cases, oocyte retrieval was performed using the long protocol of ovarian stimulation. All embryos used underwent conventional in vitro fertilization (cIVF).
Embryos were transferred to Sequential Blast culture medium (Origio, Måløv, Denmark) to be recovered for 24 hours until biopsy.

| Trophectoderm biopsy and sample stage
Trophectoderm biopsy was performed in 14 (6 dpf) and 15 (7 dpf) blastocysts. A biopsy pipette (Biopsy Flat 30°, Origio) was used to obtain 5 to 10 TE cells by suction. The biopsy pipette was then

| Next-generation sequencing
Cell processing, lysis, DNA extraction, and whole genome amplification of samples were performed using a SurePlex DNA Amplification System (Illumina, San Diego, CA, USA). Following whole genome amplification pretreatment, samples were further treated using the VeriSeq PGS kit (Illumina) and then were subjected to next-generation sequencing (NGS) analysis in a MiSeq system (Illumina). Fiorentino et al have described the protocol for using the VeriSeq PGS kit in detail. 18 To ensure the reliability of our NGS analysis, we confirmed that the values of the parameters, overall noise, the number of total reads, and the number of reads after filtering were inside the spectrum of permitted values provided in the instruction manual of the VeriSeq PGS kit. To evaluate the concordance between the chromosomal status of the biopsy sample and the remaining blastocysts, we analyzed the obtained data using the Bluefuse Multi Software. All cases except those found to be euploid were considered abnormal. Among the noted abnormalities, only those in which 100% frequency of monosomy or trisomy was observed were judged to be aneuploid, whereas those showing less than 100% increase or decrease in the number of chromosomes were judged to be mosaic. The mosaic frequency rates of cell masses analyzed using the VeriSeq PGS kit and NGS were found to be well-correlated with the analysis results. 19 Thus, when the mosaic frequency rate was found to differ, the result of our analysis was considered as discordant.

| Statistical analyses
All statistical analyses were performed using the R Project for statistical calculations (R ver. 4.0.3, Vienne, Austria). The Welch's t test was used to compare means of age, infertility, and number of oocyte retrievals. Fisher's exact test was used to compare the chromosomal status concordance between whole blastocysts with normal and abnormal karyotypes according to the TE biopsy. The same test was also used to compare the concordance between the chromosomal status of whole blastocysts with total aneuploidy and those with mosaic or other chromosomal status among groups with an abnormal chromosomal status. For three-group comparisons, the data were analyzed using the Benjamini and Hochberg method. A P-value of <.05 was considered statistically significant.

| RE SULTS
We analyzed the biopsied TE samples and remaining blastocysts using NGS and then compared their chromosomal status. The results for all samples are presented in Table 1. We found that among the 29 blastocysts, the chromosomal status of 18 (62.1%) samples was in concordance, whereas that of 11 (37.9%) blastocysts was discordant, when the 2 groups were compared.
In this study, we classified the types of concordance or discordance between the chromosomal status of biopsied TE cells and remaining blastocyst cells into 6 types. For concordance: (1) both TE and remaining blastocysts were euploid (n = 11); (2) both TE and remaining blastocysts were aneuploid (n = 7). For discordance: (3) (Table 3). Moreover, we observed that among the embryos for which a chromosomal status abnormality was identified in biopsied TE cells, all cases diagnosed as aneuploid with no mosaicism were in concordance with the chromosomal status of the whole embryo. However, we observed discordance between the wholeembryo chromosomal status and the TE biopsy results in all cases with detected abnormalities such as mosaicism, a combination of mosaic, and any other chromosomal aneuploidy, and complex chromosomal abnormalities such as a partial deletion or duplication (Table 4).
In terms of concordance according to the developmental stage of the blastocysts during biopsy, we found that the rates for unhatched, hatching, and hatched stages were 46.2% (4/13), 71.4% (5/7), and 77.8% (6/9), respectively. Although the differences were not significant, we noticed that the concordance rates were marginally matched as the developmental stage progressed (Table 5).
Regarding the patient age during oocyte retrieval, we observed that the mean age was 35.1 ± 2.7 years in the concordance group and 33.9 ± 2.8 years in the discordance group; no significant difference in age was observed between the 2 groups (P =.2685, Welch's t test). In addition, we did not observe any significant difference between embryos biopsied at 6 dpf and 7 dpf, of which the concordance rates were 57.1% (8/14) and 60.0% (9/15), respectively (P = 1.0000, Fisher's exact test).   Developing a fundamental method to solve the issue of discordance in the current PGT-A, which is performed using just a small number of cells from the embryo, is considered to be essentially impossible. Increasing the number of biopsied cells could lower the inconsistency rates but involves a higher degree of invasiveness.

| D ISCUSS I ON
Moreover, no benefits would come from increasing the number of biopsied cells, as the process would inevitably bring several negative effects such as embryo death, lower implantation rates, and higher miscarriage rates.
In our study, the concordance rate was high when no mosaicism was detected by TE biopsy, suggesting that TE biopsy is a reliable method for assessing the chromosomal status of the whole embryo. In cases of mosaicism, the consistency rates when considering the mosaic frequency were lower but could be considered high when cells were classified according to the PGDIS guidelines. However, these consistency rates were merely based on the classification criteria and did not strictly represent the frequency of abnormal cells in the actual embryo. Using the current PGT-A method, it is difficult to accurately determine the actual chromosomal status of the whole embryo.
Therefore, the development of a new technology that could replace or complement the results of analysis performed using PGT-A, such as using cfDNA present in the culture supernatant, 14-16 is warranted.

ACK N OWLED G M ENTS
We would like to thank Editage [http://www.edita ge.com] for English language editing.

D I SCLOS U R E S
Conflict of interest: The authors declare that they have no conflict of interest.

E TH I C A L A PPROVA L
The present study was conducted with the individual consent of each patient following approval by our Institution (Permission number: 1090.2) and the Ethics Committee of the Japan Society of Obstetrics and Gynecology (Permission number: 127). All patients donated their embryos after providing informed consent following the approval of the study protocol. Additionally, the embryos provided for this study were handled according to the regulations of the Japan Society of Obstetrics and Gynecology regarding research using human sperm, ova, and fertilized ova.