The present study suggested that embryos conceived after GnRH antagonists may have higher early pregnancy loss rates. This study also demonstrated that the GnRH-ant protocol was associated with a higher embryo aneuploidy rate than the GnRH-a long protocol in Chinese women receiving PGS cycles. Additionally, the rate of aneuploidy in early aborted tissues in patients who conceived after receiving the GnRH-ant protocol was significantly increased compared with that in patients who conceived after receiving the GnRH-a long protocol. Furthermore, this study showed that the embryo aneuploidy rate in the GnRH-ant protocol was significantly higher than that in the GnRH-a long protocol but only in young and normal ovarian responders. These results suggest that an increased embryo aneuploidy rate may account for the higher early pregnancy loss rate associated with GnRH-ant protocol. This is the first study that evaluated whether a difference in aneuploidy rates exists between the GnRH-ant protocol and GnRH-a long protocol.
Aneuploidy is one of the most detrimental factors causing failed implantation, miscarriage, and disordered embryo development [23]. Many researchers have explored the influencing factors of embryonic aneuploidy. Several studies in animal oocytes have suggested increased aneuploidy rates after ovulation induction [24, 25]. In humans, studies have also reported that the proportions of human preimplantation embryos with aneuploidy are increased by ovarian stimulation [26, 27]. Studies have also suggested that a higher daily dose of gonadotropin during IVF is associated with greater aneuploidy rates than a lower daily dose of gonadotropin [28, 29]. It was also found patients requiring more days of stimulation presented a lower rate of aneuploid embryos[9]. Conversely, Labarta et al. showed that ovarian stimulation did not significantly increase the embryo aneuploidy rate in IVF-derived human embryos compared with an unstimulated cycle (24). Three retrospective studies of PGS outcome data showed that the total dose of exogenous gonadotropins was not significantly associated with blastocyst aneuploidy and that a high dosage of gonadotropin did not affect euploidy or pregnancy rates [30–32]. A retrospective cohort study that included 2230 embryos conceived from IVF that underwent PGT-A also demonstrated that the gonadotropin dosage, duration of ovarian stimulation, estradiol level, follicle size at ovulation trigger and number of oocytes retrieved, within certain ranges, do not appear to significantly influence euploidy rates, regardless of the woman’s age [33]. Therefore, the potential deleterious effect of ovarian stimulation on oocyte and embryo euploidy is still the subject of lively debate. The conflicting results may be attributed to the operation of different physicians in different IVF centers. It has been found that euploidy rates for embryos created using donor oocytes can vary significantly between different IVF centers and even between donors treated by different physicians at the same IVF center [26, 34]. Studies have also demonstrated different mosaicism rates between IVF centers, implicating differences in stimulation protocols as a potential reason. Therefore, in the present study, we compared the effects of the GnRH-ant protocol and GnRH-a long protocol on embryo aneuploidy.
It has been reported that in a general IVF population, GnRH-ant protocol is associated with lower ongoing pregnancy rates than the GnRH-a long protocol [14, 35]. Studies have also suggested that embryos conceived after receiving GnRH antagonists may be associated with higher early pregnancy loss rates [15–17]. It is evident that aneuploidy is a major factor that results in early pregnancy loss [19]. However, the rate of aneuploidy in early aborted tissues had not been evaluated in patients who conceived after receiving the GnRH-ant or GnRH-a long protocol. In the present study, we noted a difference in aneuploidy rates between our two cohorts and found that the GnRH-ant protocol was associated with a significantly higher aneuploidy rate. Univariable analysis revealed that female age, male age, the basal FSH level, the basal E2 level, infertility diagnosis, and the stimulation protocol were associated with the aneuploidy rate in aborted tissues. Furthermore, after regression analysis controlling for age, the basal FSH level, the basal E2 level, infertility diagnosis, the total dose of Gn and the duration of Gn, this difference in aneuploidy rates between the GnRH-ant and GnRH-a long protocol groups was still statistically significant. The present study also showed that the GnRH-ant stimulation protocol mainly caused an increased incidence of trisomy 13, 18, and 21 compared with the GnRH agonist stimulation protocol. An increase in the rate of aneuploidy in early aborted tissues may account for the higher early pregnancy loss rates of women who conceived after receiving GnRH antagonists. Thus, it is believed that blastocysts treated with GnRH-ant protocol may exhibit more aneuploidy than blastocysts treated with GnRH-a long protocol.
Based on this hypothesis, we compared the rates of aneuploidy in blastocysts treated with GnRH antagonists and long agonists in PGS-A cycles. Given the potential variation in different PGT-A testing platforms, we evaluated blastocysts using NGS only [34]. The rate of blastocyst aneuploidy was significantly higher in the GnRH-ant protocol group. In the multivariable model, when maternal age, paternal age and potential influencing factors, including infertility diagnosis, PGT indications, the basal FSH level, the basal E2 level, the total dose of Gn and the duration of Gn, were considered, the GnRH-ant regimen remained significantly associated with an increased risk of blastocyst aneuploidy in women receiving PGT cycles. Previous studies reported that embryonic aneuploidy rates were not influenced by the dose of gonadotropins used in ovarian stimulation [30, 31]. Our study results reinforce the idea of their findings, showing that blastocyst aneuploidy is independent of the dose of gonadotropins in both the GnRH-ant and GnRH-a long protocols. Previous data have indicated that the aneuploidy rates in embryos produced from eggs collected from ovarian stimulation are between 39.1 and 53.2%, which are higher than the proportion of eggs with abnormal chromosomes in young women during the natural cycle (17%) [4, 36]. Our present data indicate that the aneuploidy rate in embryos produced from the GnRH-ant protocol is 58.00%.
The effect of the hormones administered during stimulation at the cellular level is still unknown. It has been reported that mitochondrial dysfunction that causes a decrease in ATP and/or an increase in reactive oxygen species (ROS) is sufficient to disrupt meiotic spindles [37, 38]. The percentage of mitochondria that were vacuolated in oocytes was significantly increased after ovarian stimulation in mice [39]. Moreover, the ratio of activated mitochondria to inactivated mitochondria and ATP synthesis in mouse oocytes decreased after ovarian stimulation [39]. Evidence has also demonstrated that repeated superovulation has adverse effects on the mitochondrial function of cumulus cells in rhesus monkeys or mice [40, 41]. Evidence has also indicated that repeated superovulation induces oxidative stress by elevating ROS levels in oocytes [42]. A significant decrease in ATP generation and increase in ROS caused by ovarian stimulation may adversely affect chromosomal segregation and meiotic spindle adjunction, which result in embryo aneuploidy [43, 44]. It has been reported that the use of antagonists may present an endocrinologically unfavorable scenario in which the suppression of endogenous pituitary gonadotrophin secretion may be insufficient [14]. The nonphysiological microenvironment caused by the GnRH-ant may result not only in abnormal follicular fluid biochemistry but also aberrant oocyte cytoplasmic development, which compromises mitochondrial function. However, no report has compared the effects of the GnRH-ant and GnRH-a long protocols on mitochondria and metabolism in human oocytes. In the future, it will be necessary to analyze the effects of GnRH antagonists on oocyte mitochondria and ATP synthesis.
Our study has several strengths and a few limitations. Strengths included its relatively large sample size from a single state over several years, which included 578 early miscarriage patients and a total of 466 cycles after receiving the GnRH-ant or GnRH-a long protocol. We controlled for maternal age, paternal age, infertility diagnosis, PGT indications, basal FSH level, basal E2 level, and total dose of Gn and duration of Gn were confounders in the study.
This study has some limitations. First, it was retrospective in nature, and some key statistical parameters may not have been calculated. We are aware that differences in baseline characteristics exist between patients who received the GnRH-ant and GnRH-a long protocols and that a greater proportion of low responders would be expected to receive the GnRH-ant protocol. Therefore, stratification and multiple linear regression were used to control for these parameters, most notably age. To control for age, it would be best to compare patients of similar age and ovarian reserve who receive the long agonist or GnRH-ant protocol. Second, the analyses performed in the study showed associations between an increased rate of embryonic aneuploidy and the GnRH-ant protocol, but they did not establish causality. In addition, this study was performed at a single IVF center, which may limit its generalizability. As euploidy rates for embryos created using donor oocytes can vary significantly between different IVF centers and even between donors treated by different physicians at the same IVF center [26, 34], a multicenter, randomized controlled trial would be the optimal strategy to confirm these results.