Evaluation of chromosomal abnormalities from preimplantation genetic testing to the reproductive outcomes: a comparison between three different structural rearrangements based on next-generation sequencing

Abstract

Purpose

The aim of this study was to determine factors affecting the chromosome imbalance in blastocysts and reproductive outcomes by a comparison between the reciprocal translocation (REC), inversion (INV), and Robertsonian translocation (ROB) carriers.

Methods

Couples with one partner carrying translocation or inversion underwent preimplantation genetic testing for chromosomal structural rearrangement (PGT-SR) cycles, including 215 PGT-SR cycles performed in subsequent 164 frozen-thawed embryo transfer cycles and 61 prenatal diagnoses of fetuses and 59 normal live birth babies. A total of 899 samples were processed by whole-genome amplification followed by next-generation sequencing (NGS). Karyotype and chromosome microarray analyses were used to confirm the PGT results from the amniotic fluid samples.

Results

A total of 843 blastocysts from 124 REC, 21 INV, and 35 ROB carriers were diagnosed by PGT-SR. The percentage of unbalanced blastocysts was significantly higher in REC than in INV and ROB carriers (64.31% vs. 28.05% vs. 37.02%). Stratification analysis of female carrier age and gonadotropin doses showed no significant increase in unbalanced chromosomal abnormalities in the three groups. Also, the different breakpoints in chromosomal arms did not affect the rate of unbalanced chromosomes in the embryos. Logistic regression indicated blastocyst quality as a statistically significant risk factor associated with unbalanced chromosomal abnormalities from translocation carriers (P < 0.001). The source of abnormalities in the three groups showed significant differences such that the abnormalities in REC mostly originated from parental translocation but the abnormalities in INV were mainly de novo variations. 164 blastocysts were transferred, and there were no significant differences in the clinical pregnancy rate and miscarriage rate. A total of 59 healthy babies were born, and there were no significant differences in the gender ratio and birth height, except the birth weight of boys between INV and ROB groups (P = 0.02). The results of amniocentesis revealed that more fetuses have normal chromosomal karyotypes than balanced carriers, particularly in the REC group.

Conclusions

Reciprocal translocation carriers have more risk of unbalanced rearrangement, but embryonic chromosome abnormalities of inversion carriers come mainly from de novo variations. This is the first study specifically comparing three different PGT-SRs using the NGS method and evaluating their reproductive outcomes. Our findings will provide the reciprocal translocation, inversion, and Robertsonian translocation carrier couples with more accurate genetic counseling on the reproductive risk of chromosomal imbalance.

Supplementary Information

The online version contains supplementary material available at 10.1007/s10815-020-02053-5.

Introduction

Balanced chromosome rearrangements are the most common chromosomal structural abnormalities in the general population, including reciprocal translocations (REC), inversions (INV), and Robertsonian translocations (ROB). However, balanced chromosome rearrangements may segregate in a balanced or an unbalanced manner. The balanced manner being neutral rarely develops a clinical phenotype, unless key genes are disrupted at breakpoints, while the unbalanced manner may result in infertility, recurrent miscarriages or offspring with developmental delay, intellectual disability, and birth defects [1]. Balanced chromosome rearrangements are not very rare, and its prevalence in prenatal samples and newborns is 0.40% and 0.19%, respectively [2].

In general, couples with chromosomal structural rearrangements undergoing natural conception have a risk (50% or more) of chromosomal abnormalities, which can increase the miscarriage rate and decrease the live birth rate [3]. Preimplantation genetic testing for structural rearrangements (PGT-SR), a feasible technique, can decrease the miscarriage rate and improve reproductive outcomes compared with the naturally conceived pregnancies in these couples [4, 5]. Although embryonic chromosomal abnormalities of PGT-SR are extensively studied, particularly in reciprocal translocations affected by interchromosomal effect (ICE) [6–8], the comparison between the three common structural rearrangements (REC, INV, and ROB) in PGT is rarely reported. Barring the controversial impact of ICE, the identification of other factors affecting chromosomal abnormalities, such as the female age, embryonic quality, and breakpoint, is still questionable [6–8]. On the contrary, gonadotropin doses reportedly increased the aneuploidy rate exclusively in the women undergoing in vitro fertilization (IVF), but other studies believed not [9–11]. Numerous studies investigating whether the gender of the carrier undergoing PGT affects the aneuploidy rate have led to inconsistent conclusions [6, 7]. Irrespective of the inheritance of the parental structural arrangement, couples with normal and balanced chromosome rearrangements produce embryos with de novo chromosomal abnormalities [2, 12]. However, the comparison of the embryonic de novo chromosomal abnormalities was uncertain between the three groups.

This study was aimed at evaluating various factors, including female age, carrier’s gender, gonadotropin doses, blastocyst score, and breakpoint in the chromosome arms affecting chromosomal abnormalities (aneuploidy and segmental anomalies) of the blastocyst, with further comparison between the REC, INV, and ROB carriers using next-generation sequencing (NGS) technology. Besides, the origin of chromosomal abnormalities in the embryos was also analyzed by observing parental and de novo abnormalities. Furthermore, the reproductive outcomes including the results of prenatal diagnosis were evaluated and compared between the three PGT-SRs.

To our knowledge, this is the first systematic evaluation and comparison between the three different PGT-SRs using the NGS method, including their reproductive outcomes.

Materials and methods

Study population

The study was approved by the Ethics Committee of Reproductive Medicine Study at Sun Yat-sen Memorial Hospital of Sun Yat-sen University, and all the patients underwent genetic counseling and signed a consent form approved by the local ethics committee. This retrospective study included 215 PGT-SR cycles, and a total of 180 couples who had blastocysts biopsied between January 2017 and December 2019 were evaluated (Table 1). No biopsied blastocysts were excluded. The chromosomal reciprocal translocation, inversion, and Robertsonian translocation status were confirmed by karyotype analysis from couples who had a history of infertility, recurrent spontaneous abortion, or pregnancy with a chromosomally abnormal fetus. For de novo chromosomal abnormality analysis, 76 age-matched patients with normal karyotype as the control group were recruited.

Table 1.

The clinical characteristics of patients who underwent PGT-SR

Characteristic	Reciprocal translocation	Inversion	Robertsonian translocation	Total
No. of cycles	150	23	42	215
No. of patients	124	21	35	180
Female age (y, mean ± SD)	30.54 ± 3.60	31.96 ± 3.64	29.33 ± 3.76	30.46 ± 3.69
Initiation Gn dose (IU, mean ± SD)	200.08 ± 52.55	213.04 ± 49.50	193.75 ± 56.26	200.23 ± 52.98
Total Gn dose (IU, mean ± SD)	2328.50 ± 692.40	2311.41 ± 461.90	2237.80 ± 826.01	2308.95 ± 698.37
Total collected oocytes (mean ± SD)	2507 (16.71 ± 9.94)	268 (11.65 ± 5.67)	709 (16.88 ± 8.72)	3484 (16.20 ± 9.45)
Total injected oocytes (mean ± SD)	1988 (13.25 ± 8.01)	203 (8.83 ± 4.55)	532 (12.67 ± 6.82)	2723 (12.67 ± 7.58)
Zygotes (2PN) (mean ± SD)	1581 (10.54 ± 6.38)	178 (7.73 ± 4.00)	411 (9.79 ± 4.80)	2170 (10.09 ± 5.93)
Blastocyst culture (mean ± SD)	1480 (9.87 ± 6.17)	171 (7.43 ± 4.26)	379 (9.02 ± 4.87)	2030 (9.44 ± 5.79)
Blastocysts biopsied (mean ± SD)	615 (4.10 ± 2.55)	90 (3.91 ± 2.78)	194 (4.62 ± 3.43)	899 (4.18 ± 2.76)
No. of diagnosed blastocysts	580	82	181	843
No. of balanced blastocysts (%)	207 (35.69)	59 (71.95)	114 (62.98)	380 (45.08)
No. of unbalanced blastocysts (%)	373 (64.31)a	23 (28.05)a	67 (37.02)a	463 (54.92)
No. of blastocysts without result	35	8	13	56
Available blastocysts (mean ± SD)	1.37 ± 1.38	2.61 ± 2.13	2.71 ± 2.27	1.77 ± 1.77

2PN two pronuclear, Gn gonadotropins, PGT-SR preimplantation genetic testing for structural rearrangement

^aP < 0.001

All the pregnant patients were followed up till parturition, and subsequently, the frozen-thawed embryos were subjected to frozen-thawed embryo transfer (FET) to confirm the results of prenatal diagnosis and observe whether any pregnancy loss had occurred.

Ovarian stimulation, intracytoplasmic sperm injection (ICSI), embryo culture, and biopsy

All patients underwent controlled ovarian hyperstimulation (COH) using recombinant follicle-stimulating hormone (FSH) (Gonal-F®, Merck Serono, Germany) and human menopausal gonadotropin (Menopur®, Ferring, Germany) by following the gonadotropin-releasing hormone (GnRH) agonist long protocol or GnRH antagonist protocol based on the ovarian reserve. Human chorionic gonadotropin (hCG) (Ovidrel®, Merck Serono, Germany) was administered as a trigger when at least 2 follicles had reached 18 mm. Transvaginal follicular aspiration was performed 36 h later. ICSI was administered on metaphase II (MII) oocytes. The oocytes were assessed for fertilization at approximately 18 h postinjection by checking the presence of two pronuclei and the second polar body. Embryos were cultured in G1-plus/G2-plus (Vitrolife, Sweden) sequential media in a humidified atmosphere containing 5% O₂ and 6% CO₂.

According to the Gardner grading system [13], blastocysts graded as 3-6, with either of the inner cell mass (ICM) or trophectoderm (TE) graded above C, were considered suitable blastocysts for biopsy. If the score of ICM and TE was simultaneously equal to or greater than B, this blastocyst was believed as a high-quality embryo. TE biopsy was performed on day 5 or day 6 by zona drilling with a laser, and a few biopsied TE cells (5–8 cells) were removed and collected for genetic analysis [14]. A single blastocyst was vitrified in sequence after the biopsy following the manufacturer’s protocols (Kitazato Corporation, Japan).

Whole-genome amplification and next-generation sequencing

Biopsied TE cells were used for whole-genome amplification with the WGA4 GenomePlex Single Cell Whole Genome Amplification Kit (Sigma-Aldrich, USA) [15]. NGS used for comprehensive chromosomal screening was performed on an MGISEQ-500 sequencer (MGI, China) using an MGICare Detection Kit for Single Cell Chromosome Copy Number Variation Test v2.0 (MGI) and dual-index single-end 35 bp reads. The results were analyzed by BGI (Shenzhen, China) and aligned to the hg19 human reference genome (http://hgdownload.cse.ucsc.edu/downloads.html). Chromosomal aneuploidy and copy numbers of chromosomal segments larger than 4 Mb can be reliably predicted with the PSCC algorithm [16, 17].

Embryo transfer and prenatal and postnatal follow-up

Only one vitrified-warmed balanced blastocyst was selected for transfer. Serum β-hCG levels were evaluated 14 days after FET. The presence of a gestational sac, detected by ultrasonography, about 4 weeks after FET was defined as clinical pregnancy. Amniocentesis was performed at about 18–24 weeks of gestation. Karyotyping and chromosome microarray analyses were used to confirm the PGT results by detecting the amniotic fluid samples. Data on pregnancy loss and delivery were acquired through a telephonic interview.

Statistical analysis

All statistical procedures were conducted with SPSS version 23.0 software. Data are presented as mean ± SD for continuous variables, and categorical variables were summarized by frequency and percentage. The correlations between variables were assessed by Spearman’s tests. Comparison between groups was performed using the chi-square test and Fisher’s exact test for these categorical variables. One-way analysis of variance was used to assess the difference between groups for continuous variables, and the least significant difference test was used for the comparison between two groups. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated by univariate logistic regression. All statistical tests were two-tailed and a P value < 0.05 was considered statistically significant.

Results

Clinical characteristics

A total of 215 PGT-SR cycles were included in the study. Among the 180 couples, 124 couples carried REC, 21 couples carried INV, and 35 couples carried ROB. The average age of females was 30.46 years. The mean initiation and total gonadotropin dose were 200.23 IU and 2308.95 IU, respectively. A total of 3484 oocytes were retrieved, 2723 MII oocytes underwent ICSI, and 843 blastocysts on day5 or day6 were successfully diagnosed (Table 1). The rate of unbalanced blastocysts showed significant differences in each group (P < 0.001), but it was not related to female age, initiation, or total gonadotropin dose (P > 0.05 for all comparisons) (Table 1). In addition, the available blastocyst rate (defined as the percentage of balanced blastocysts among biopsied blastocysts) was 45.08% in PGT-SR, among which the rate of the inversion group (71.95%) was highest. The mean number of available blastocysts in PGT-SR was 1.77 per cycle (Table 1).

Chromosomal abnormality analysis of PGT for structural rearrangement embryos

To evaluate the effects of the carrier’s gender and female age on unbalanced embryos in different PGT-SRs, the female age groups were further subdivided for analysis into two categories: those aged < 35 years and those aged ≥ 35 years. As shown in Table 2, there were no significant differences in the unbalanced embryo rate in each group (P > 0.05 for all comparisons) demonstrating that the rate of occurrence of the abnormal embryo is not significantly influenced by the carrier’s gender and female age.

Table 2.

Analysis of the effect of the carrier’s gender and female age on unbalanced embryos in different PGT-SRs

Variable		Balanced, n (%)	Unbalanced, n (%)	P	OR (95% CI)
Reciprocal translocation
Carrier’s gender	Female	122 (35.47)	222 (64.53)	0.892	0.976 (0.691-1.379)
	Male	85 (36.02)	151 (63.98)
Female age	≥ 35 years	15 (31.25)	33 (68.75)	0.503	0.805 (0.426-1.520)
	< 35 years	192 (36.09)	340 (63.91)
Inversion
Carrier’s gender	Female	29 (74.36)	10 (25.64)	0.644	1.257 (0.477-3.314)
	Male	30 (69.77)	13 (30.23)
Female age	≥ 35 years	4 (66.67)	2 (33.33)	0.765	0.764 (0.130-4.484)
	< 35 years	55 (72.37)	21 (27.63)
Robertsonian translocation
Carrier’s gender	Female	47 (57.32)	35 (42.68)	0.151	0.641 (0.349-1.177)
	Male	67 (67.68)	32 (32.32)
Female age	≥ 35 years	9 (64.29)	5 (35.71)	0.916	1.063 (0.341-3.315)
	< 35 years	105 (62.87)	62 (37.13)

PGT-SR preimplantation genetic testing for structural rearrangement

To evaluate the effect of the blastocyst score on unbalanced embryos in different PGT-SRs, the blastocyst score groups were subdivided into two categories: the score of ICM and TE ≥ BB (high-quality embryo) and the score of ICM and TE < BB, but > CC (low-quality embryo). The percentage of unbalanced blastocysts was higher in the low-quality embryo than in the high-quality embryo in REC carriers (71.05% vs. 54.62%) with a significant statistical difference (P < 0.001, OR 2.039, 95% CI 1.443–2.882). Moreover, the percentage of unbalanced blastocysts was higher in the low-quality embryo in ROB carriers (46.81% vs. 26.44%) with a statistical difference (P = 0.005, OR 2.449, 95% CI 1.310–4.577). However, there were no statistically significant differences in the INV group (Fig. 1) (Table 3).

Fig. 1.

Percentage of balanced and unbalanced blastocysts of the different scores in the three PGT-SR groups. a The score of ICM and TE ≥ BB (high-quality embryo). b The score of ICM and TE < BB, but > CC (low-quality embryo). The light blue bar indicates the percentage of balanced blastocysts; the red bar indicates the percentage of unbalanced blastocysts. REC reciprocal translocation group, INV inversion group, ROB Robertsonian translocation group

Table 3.

Analysis of the effect of the blastocyst score on unbalanced embryos in different PGT-SRs

Variable		Balanced, n (%)	Unbalanced, n (%)	P	OR (95% CI)
Reciprocal translocation
Blastocyst score	≥ BB	108 (45.38)	130 (54.62)	< 0.001a	2.039 (1.443-2.882)
	< BB	99 (28.95)	243 (71.05)
Inversion
Blastocyst score	≥ BB	38 (77.55)	11 (22.45)	0.169	1.974 (0.744-5.241)
	< BB	21 (63.64)	12 (36.36)
Robertsonian translocation
Blastocyst score	≥ BB	64 (73.56)	23 (26.44)	0.005a	2.449 (1.310-4.577)
	< BB	50 (53.19)	44 (46.81)

BB blastocysts graded as 3–6 and both of the inner cell mass (ICM) and trophectoderm (TE) graded as B, PGT-SR preimplantation genetic testing for structural rearrangement

^aStatistically significant

To evaluate the effect of chromosome arms with a breakpoint on unbalanced embryos in the REC group and the INV group, we divided the chromosome arms into three subgroups: both breakpoints in the short arms of the chromosome (pp), both breakpoints in the long arms of the chromosome (qq), and one breakpoint in the short arm and another breakpoint in the long arm (pq or qp). But no significant differences were observed between any of the groups demonstrating that the unbalanced embryo rate is not significantly impacted by the position of breakpoints (Fig. 2).

Fig. 2.

Percentage of balanced and unbalanced blastocysts with different breakpoints in the chromosome arms. a Reciprocal translocation group. b Inversion group. The light blue bar indicates the percentage of balanced blastocysts; the red bar indicates the percentage of unbalanced blastocysts. pp both breakpoints in the short arms of the chromosome, qq both breakpoints in the long arms of the chromosome, pq+qp one breakpoint in the short arm and another in the long arm

To evaluate the chromosomal abnormality origin of embryos in different PGT-SRs, abnormal patterns were subdivided into three categories: the parental group comprising embryonic chromosomal abnormalities associated with paternal or maternal balanced translocation carriers, the de novo group comprising novel chromosomal abnormalities, excluding paternal or maternal origin, and the parental+de novo group comprising both of the above abnormalities. In the REC group, 63.81% chromosomal abnormality originated from parental balanced translocation, but in the INV group, 82.61% chromosomal abnormality was derived from de novo variations (Supplemental Table 1). Significant differences were observed between the groups (P < 0.001) (Fig. 3).

Fig. 3.

Percentage of the chromosomal abnormality origin of abnormal blastocysts between the three PGT-SR groups. The parental group comprised an embryo with chromosomal abnormalities associated with paternal or maternal balanced translocation carriers, the de novo group comprised novel chromosomal abnormalities, excluding paternal or maternal origin, and the parental+de novo group comprised both the above abnormalities. REC reciprocal translocation group, INV inversion group, ROB Robertsonian translocation group

To evaluate the de novo chromosomal abnormalities in different PGT-SRs and normal controls, we divided the chromosomes into two subgroups: the de novo group comprising novel chromosomal abnormalities (including aneuploidy and segmental anomalies) excluding paternal or maternal origin and the balanced and parental group comprising normal or balanced translocation embryonic chromosomes and embryo chromosomal abnormalities associated with paternal or maternal balanced translocation carriers. In the REC group, a total of 580 (26,680 chromosomes) blastocysts were detected, and 178 de novo chromosomal abnormalities from 135 blastocysts were identified. The average de novo variant probability of a given chromosome was 0.87% (178/26,680). In the INV and ROB groups, 0.56% (21/3,772) and 0.54% (45/8,326) for de novo variants per chromosome were identified, respectively. In the control group, among 249 (11,454 chromosomes) blastocysts, 73 de novo chromosomal abnormalities from 54 blastocysts were assessed, indicating the average de novo variant probability per chromosome being 0.64%. But no significant differences were observed between any of the groups (Supplemental Table 2).

Clinical outcomes and prenatal diagnosis

In FET cycles, a total of 98, 22, and 44 available blastocysts were transferred in the REC, INV, and ROB groups, respectively, and a single blastocyst was transferred to each patient. A total of 59 healthy babies were born including 57 singletons and a set of twins, while 14 couples had an ongoing pregnancy (Table 4). There were no significant differences in the biochemical pregnancy rate and clinical pregnancy rate per transfer between the three groups. No significant differences were also observed in the miscarriage rate per clinical pregnancy between the three groups (P = 0.954). The gender ratio at birth did not differ for patients in the three groups (P = 0.058). Moreover, there were no differences in birth height and birth weight of girls between the three groups, but the birth weight of boys in the ROB group was lower than the birth weight of boys in the INV group (P = 0.02) (Table 4).

Table 4.

The clinical outcomes of patients who underwent FET

Characteristic	Reciprocal translocation	Inversion	Robertsonian translocation	Total
ET cycles	98	22	44	164
Transferred blastocyst number	98	22	44	164
Biochemical pregnancies (per ET cycle)	12 (12.24)	4 (18.18)	5 (11.36)	21 (12.80)
Clinical pregnancies (per ET cycle)	53 (54.08)	13 (59.09)	21 (47.73)	87 (53.05)
Twin pregnancies (per clinical pregnancy)	1 (1.89)	0 (0)	0 (0)	1 (1.15)
Miscarriage (per clinical pregnancy)	7 (13.21)	2 (15.38)	4 (19.05)	13 (14.94)
Ectopic pregnancies (per clinical pregnancy)	0 (0)	1 (7.69)	0 (0)	1 (1.15)
Ongoing pregnancy	9	2	2	13
Delivered babies	36	8	15	59
No. of boys	25	3	6	34
No. of girls	11	5	9	25
Gestational age of boys (weeks, mean ± SD)	38.52 ± 1.26	39.50 ± 1.80	38.50 ± 1.20	38.61 ± 1.29
Gestational age of girls (weeks, mean ± SD)	39.68 ± 0.94	38.94 ± 0.75	39.14 ± 1.18	39.34 ± 1.02
Birth height of boys (cm, mean ± SD)	49.67 ± 1.83	51.67 ± 1.53	49.00 ± 2.00	49.73 ± 1.91
Birth height of girls (cm, mean ± SD)	50.00 ± 1.55	48.80 ± 2.68	49.25 ± 1.16	49.50 ± 1.72
Birth weight of boys (g, mean ± SD)	3193.20 ± 434.92	3700.00 ± 427.72a	2798.33 ± 435.00a	3168.24 ± 477.86
Birth weight of girls (g, mean ± SD)	3326.36 ± 303.06	3190.00 ± 551.59	3160.00 ± 346.55	3239.20 ± 367.90

ET embryo transferred

^aP = 0.02

During our study period, 61 pregnant women were subjected to midtrimester amniocentesis tests including 61-karyotype analysis and 43-chromosome microarray analysis. We did not find any aneuploidy and deleterious segmental anomalies, and all the results of prenatal diagnosis were consistent with PGT. In the REC group, the number of fetuses with normal karyotypes was larger than the balanced carriers (23 vs. 14), but in the ROB group, the balanced carriers were more (Table 5).

Table 5.

The prenatal diagnosis results of patients who underwent PGT

Characteristic		Reciprocal translocation	Inversion	Robertsonian translocation	Total
No. of patients		37	9	15	61
Fetus karyotype	Normal	23	5	6	34
	Balanced carrier (paternal/maternal)	14	4	9	27

PGT-SR preimplantation genetic testing for structural rearrangement

Discussion

This study has comprehensively evaluated the embryonic chromosomal abnormal features from 215 PGT-SR cycles and the follow-up data of 61 prenatally diagnosed fetuses. So far, 59 healthy babies have been born and the amniocentesis results show that all fetuses have balanced or normal chromosomal karyotype. To our knowledge, this is the largest sample size used in PGT-SR cycles for clinical outcomes including prenatal diagnosis. This study has identified that low-quality blastocysts are associated with a higher incidence of chromosomal abnormalities in the REC and ROB groups. Moreover, chromosomal unbalanced rearrangements in blastocysts of inversion carriers are mostly de novo variants that are different from REC and ROB carriers, but the average de novo variant probability of a given chromosome revealed no significant differences between the three PGT-SRs and control groups.

Balanced translocations are usually without a phenotypic effect but are associated with a high risk for unbalanced gametes and abnormal progeny. During meiosis, the chromosomes of a carrier of a balanced reciprocal translocation pair must form a quadrivalent to ensure proper alignment of homologous sequences, in which the chromosomes can segregate and generate alternate, adjacent-1, adjacent-2, 3:1, and 4:0 segregation [18, 19]. Previous studies presented that a meiotic segregation pattern might be affected by the carrier’s gender, which revealed that the incidences of adjacent-1, adjacent-2, and 3:1 segregation were different in female and male carriers [7, 20, 21]. However, among these studies, the frequency of an alternate segregation pattern (normal and balanced) was no different in male versus female translocation carriers, which was consistent with our study. Our results revealed that the rates of unbalanced chromosomal abnormalities (not limited to quadrivalent origin) in blastocysts might not be affected by the carrier’s gender, but the unbalanced embryo rate was slightly increased in female carriers. Another study analyzed the carrier’s gender and found an increase in the percentage of abnormal embryos for female carriers of REC and ROB groups [6]. However, this study was restricted to day 3 blastomere embryos and the small sample size in blastocysts, but 101 diagnosed blastocysts of the REC group were not affected by the carrier’s gender [6]. In addition, we found that PGT for unbalanced inversion also was not affected by gender, including further analysis in PAI and PEI carriers (Supplemental Table 3). These findings were similar to the previous study in PAI carriers [8]. However, they found that the outcome in PEI carriers was affected by the carrier’s gender, and the meiotic segregation mechanism requires further investigation [8].

According to previous reports, advanced maternal age contributed to the increased incidence of chromosomal abnormalities [22]. Our data also revealed that the unbalanced embryo rate of the advanced maternal age group (≥ 35 years) was mildly higher than that of the younger group (< 35 years), but there were no significant differences between the REC, INV, and ROB groups (Table 2). Other studies also confirmed our above results that the frequency of abnormal embryos might not be affected by the carrier’s age [7, 20]. To further exclude the age factor, we analyzed the female age associated with the percentage of unbalanced blastocysts in the three groups, respectively, but there were no correlations between them (Table 1). We supposed that structure rearrangement, as the major factor, would affect the unbalanced status in balanced carriers, which might be different from the results obtained by normal karyotyping in PGT for aneuploidies (PGT-A). The advanced maternal age was the major factor in PGT-A patients [23]. Additionally, some researchers opine that gonadotropin doses might be associated with embryonic aneuploidy in IVF [9]. In contrast, our study showed that the initiation and total gonadotropin doses did not affect the rate of blastocyst aneuploidy, which was similar to other PGT studies [10, 24].

In addition, we observed that the rate of blastocyst biopsied in the REC group was the lowest in comparison with that in the other two groups (41.55% vs. 52.63% vs. 51.19%, P < 0.001). It implied that the rate of available blastocysts might be related to chromosomal structural rearrangements. Further, we found a higher percentage of unbalanced blastocysts in the low-quality embryo group of reciprocal translocation carriers (P < 0.001) and Robertsonian translocation carriers (P = 0.005). These results indicated that balanced blastocysts could have a higher embryonic score compared with unbalanced blastocysts. Another study compared balanced blastocyst morphology with unbalanced translocations and found significant differences in the high blastocyst grading group, corroborating our findings [25]. This phenomenon might be associated with delayed development and asynchronous cleavage by time-lapse imaging [25]. Moreover, abnormal embryos might arrest in the blastomere stage, which implied that partial embryos of developmental potentiality would enter into the blastocyst stage and were euploidy. Actually, day 6 blastocysts had longer times for every time point analyzed than day 5 blastocysts under the time-lapse morphokinetics analysis, but there was no statistically significant difference in euploidy rates of blastocysts [26]. However, chromosomal abnormalities associated with early embryonic development need further investigations.

In our study, the rate of unbalanced blastocysts in the INV group was the lowest compared with that of the REC and ROB groups (Table 1), and it was not related to the carrier’s gender and female age (Table 2). Some retrospective studies investigated the interchromosomal effects in blastocysts from inversion preimplantation genetic testing cycles [8, 27]. It was proposed that ICE played a milder effect on inversions, such that the carriers of balanced chromosome inversions do not exhibit higher aneuploidy rates than maternal age-matched controls [8, 27]. However, the ICE usage was always controversial, and it was necessary to design additional studies to identify the impact of translocation nondisjunction and the effect of chromosome exchange in both meiosis and mitosis [28]. Given the chromosomal abnormality origin, our study found that de novo variants were the major abnormalities in the INV group, which was different from the REC and ROB groups (Fig. 3) (Supplemental Table 1). Further, we evaluated the de novo chromosomal abnormalities in different PGT-SRs and maternal age-matched controls, which revealed that the average de novo variant probability of a given chromosome had no significant differences in each group. Therefore, in comparison with the REC and ROB groups, the proportion of de novo variants in the INV group seemed to be in majority. We suspected that the inversion ring might not increase genome instability during meiosis compared with REC and ROB groups. On the other hand, our results also showed that the REC group would increase the frequency of meiotic chromosome nondisjunction that originated from parental balanced translocation compared with ROB and INV groups, in accordance with the previous study [18, 28]. Moreover, previous studies showed that other factors (breakpoint and inverted segment size) might influence the generation of unbalanced rearrangement [29, 30]. Their results revealed that in inversion carriers, the unbalanced productions seemed to be directly related to the size of the inversion when the inverted segment is larger than 100 Mb [30]. However, Fig. 2 shows no significant differences in any of the groups demonstrating that the unbalanced embryo rate was not significantly affected by the position of breakpoints. Another study also analyzed the breakpoint and found that meiotic segregation might not be affected by breakpoints in reciprocal translocation and inversion carriers, corroborating our findings [6, 20]. Since we did not analyze the size of genetic distances between the two breakpoints and the association with abnormal segregation, our results reflected that the position of breakpoints was not related to the rate of the unbalanced embryo. But our results also displayed a trend towards higher percentage of chromosomal abnormalities with the pq+qp group. The breakpoints associated with segregation modes need further investigations.

The clinical outcomes revealed that PGT could decrease miscarriage rates and improve pregnancy outcomes. We did not find significant differences in clinical pregnancy rates, miscarriage rates, birth height, and birth weight between the three groups, except that the birth weight of boys in the ROB group was lower than the birth weight of boys in the INV group (Table 4). Besides, the karyotype analysis performed on all the amniotic fluid samples revealed normal or balanced carriers. In addition, fetuses with normal karyotypes were found to be higher than those with balanced translocation in the REC group, similar to other studies [31]. No aneuploidy or pathogenic copy number variations were found in chromosome microarray analysis, besides 4 cases that were believed to be benign or have uncertain significance according to the ACMG guideline [32] (Supplemental Table 4). However, we did not find any birth defects, and 59 babies were born by normal live birth.

In summary, chromosomal abnormalities were assessed in 843 blastocysts from reciprocal translocation, inversion, and Robertsonian translocation carriers. The results indicated that the rate of unbalanced chromosomal abnormality was not affected by the carrier’s gender and breakpoint, and high-quality blastocysts from translocation carriers were more likely to be balanced. There were significant differences between the unbalanced embryo rates between the three groups, and reciprocal translocations were susceptible to the unbalanced rearrangement effect, but inversions were not. The chromosomal abnormalities in blastocysts of inversion carriers mainly originated from de novo variants, and inversion carriers would have a more than 70% chance for obtaining balanced embryos. Finally, there were no significant differences in the clinical pregnancy rate, miscarriage rate, gender ratio, and birth height between the three groups. PGT-SR based on NGS is a reliable method, and the number of normal karyotypes was likely higher than that in the balanced carriers according to the results of prenatal diagnoses. The above results will help provide accurate genetic counseling to reciprocal translocation, inversion, and Robertsonian translocation carrier couples.

Funding

This work was partially supported by the National Natural Science Foundation of China (No. 81801431), the Chinese Medical Association Clinical Medical Research Special Fund, the Research and Development of Young Physicians in Reproductive Medicine (No. 18010060735), the Natural Science Foundation of Guangdong Province (No. 2019A1515012005), and the Guangzhou Science and Technology Project (No. 201704020217).