Preimplantation genetic testing for aneuploidy could not improve cumulative live birth rate among 1003 couples with recurrent pregnancy loss

To the Editor: Recurrent pregnancy loss (RPL) is defined as two or more consecutive spontaneous pregnancy losses before 24 weeks of gestation; it affects 1–2% of couples.[1] Some causes of RPL are uterine anatomical abnormalities, abnormal chromosomes in parents or embryos, antiphospholipid syndrome, and certain autoimmune disorders.[2] However, the etiology of 40–50% of RPL couples remains unclear and is defined as unexplained RPL (uRPL). Approximately 41% of pregnancy losses can be attributed to fetal aneuploidy,[3] common in couples with advanced maternal age (AMA). Although Preimplantation Genetic Testing for Aneuploidy (PGT-A) has been considered a potential technique to improve pregnancy outcomes, currently there is no consensus on its application. A recent retrospective SART-CORS study indicated that PGT-A increases the rate of live births and clinical pregnancies for frozen embryo transfer (FET) in RPL couples.[4] However, a multicenter, prospective study did not find any benefits of PGT-A in improving the live birth rate per patient or reducing the rate of miscarriage in RPL couples.[5] Nonetheless, these studies included only a single embryo transfer (ET) cycle or assessed the efficacy of PGT-A in terms of live birth rate per transfer, which does not have sufficient clinical significance. To determine whether PGT-A impact the cumulative live birth rate after in vitro fertilization (IVF) in women with RPL, we included 1003 uRPL couples with ≥21-year-old females from January 2015–2022 at the Reproductive Center of Shandong University (Jinan, Shandong, China). Patients provided informed consent and the study was approved by the Reproductive Center of Shandong University's Ethics Committee ([2021] IRB No. [140]). The sample size met design requirement. uRPL was defined as two or more consecutive miscarriages with no distinct causes. The exclusion criteria for patients were parental chromosomal abnormalities, uterine anatomical abnormalities, endocrine dysfunction, autoimmune disorders, thrombosis, and infection. The uRPL couples were divided into PGT-A and conventional IVF (control) groups according to whether they underwent PGT-A. The cumulative pregnancy outcomes after a single oocyte retrieval were followed up and analyzed. The primary outcome was cumulative live birth rate following a single oocyte retrieval. The other pregnancy outcomes included cumulative rates of clinical pregnancy, and pregnancy loss after all ET cycles following a single oocyte retrieval, rates of clinical pregnancy and pregnancy loss after the first ET. These pregnancy outcomes were defined as described previously.[6,7] Ovarian stimulation, preimplantation genetic testing for aneuploidy PGT-A, endometrial preparation luteal-phase support and statistical analysis were described in detail in the supplementary methods, https://links.lww.com/CM9/B923. A total of 1003 couples with a history of uRPL were included in this study. Among them, 799 underwent PGT-A (392 in <35 years subgroup, 407 in ≥35 years subgroup), while 204 were not treated with PGT-A and comprised the control group (142 in <35 years subgroup, 62 in ≥35 years subgroup). The baseline characteristics of RPL couples were shown in Supplementary Tables 1 and 2, https://links.lww.com/CM9/B923. The majority of them were comparable between the PGT-A group and the control group among the younger and older women. Moreover, we summarized the oocyte retrieval and embryo development protocols [Supplementary Tables 3 and 4, https://links.lww.com/CM9/B923]. We found that the cumulative rates of live birth (53.32% [209/392] vs. 61.97% [88/142], χ2 = 3.164, P = 0.077) and clinical pregnancy loss (19.64% [55/280] vs. 18.42% [21/114], χ2 = 0.078, P = 0.888) were similar between the two groups among women aged <35 years [Table 1]. Also, the cumulative live birth rate in women ≥35-year-old was similar between the two groups (28.75% [117/407] vs. 30.65% [19/62], χ2 = 0.094, P = 0.765). However, couples in the PGT-A group showed significantly lower cumulative rates of biochemical pregnancy loss (10.13% [16/158] vs. 32.56% [14/43], χ2 = 13.394, P = 0.001) and clinical pregnancy loss (20.89% [33/158] vs. 37.21% [16/43], χ2 = 4.885, P = 0.03) than those in the control group [Table 1]. Table 1 - Comparison of cumulative pregnancy outcomes between PGT-A and non-PGT-A patients with RPL undergoing FET. Characteristics RPL Statistic value P-value PGT-A Non-PGT-A <35 years No. of cases, n 392 142 Cycles of ET, n 450 221 Live births/women, n/N (%) 209/392 (53.32) 88/142 (61.97) 3.164 0.077 Biochemical pregnancies/women, n/N (%) 280/392 (71.43) 114/142 (80.28) 4.224 0.045* Clinical pregnancies/women, n/N (%) 253/392 (64.54) 106/142 (74.65) 4.833 0.029* Ongoing pregnancies/women, n/N (%) 218/392 (55.61) 94/142 (66.20) 4.808 0.029* Pregnancy loss/biochemical pregnancies, n/N (%) Biochemical 41/280 (14.64) 18/114 (15.79) 0.084 0.876 Clinical 55/280 (19.64) 21/114 (18.42) 0.078 0.888 First 42/280 (15.00) 16/114 (14.04) 0.060 0.876 Second 14/280 (5.00) 5/114 (4.39) 0.067 1.000 ≥35 years No. of cases, n 407 62 Cycles of ET, n 260 99 Live births/women, n/N (%) 117/407 (28.75) 19/62 (30.65) 0.094 0.765 Biochemical pregnancies/women, n/N (%) 158/407 (38.82) 43/62 (69.35) 20.483 <0.001* Clinical pregnancies/women, n/N (%) 148/407 (36.36) 31/62 (50.00) 4.239 0.049* Ongoing pregnancies/women, n/N (%) 127/407 (31.20) 20/62 (32.26) 0.028 0.884 Pregnancy loss/biochemical pregnancies, n/N (%) Biochemical 16/158 (10.13) 14/43 (32.56) 13.394 0.001* Clinical 33/158 (20.89) 16/43 (37.21) 4.885 0.030* First 24/158 (15.19) 11/43 (25.58) 2.538 0.118 Second 9/158 (5.70) 4/43 (9.30) – 0.482 *P <0.05 represents significance difference. ET: Embryo transfer; FET: Frozen embryo transfer; PGT-A: Pre-implantation genetic testing for aneuploidy; RPL: Recurrent pregnancy loss; –: No statistic value with Fisher's exact test. We also analyzed pregnancy outcomes after the first ET. The results did not differ significantly between the two groups in younger women. Interestingly, a lower biochemical pregnancy rate (33.66% [137/407] vs. 50.00% [31/62], χ2 = 6.248, P = 0.015) and a higher biochemical pregnancy loss rate (8.03% [11/137] vs. 35.48% [11/31], P <0.001) were detected in the older patients of the PGT-A group [Supplementary Table 5, https://links.lww.com/CM9/B923]. In addition, case-control matching for baseline characteristics, results of oocyte retrieval and embryo culture of PGT-A and non-PGT-A patients was performed [Supplementary Tables 6–9, https://links.lww.com/CM9/B923]. After case-control matching, the pregnancy outcomes, especially the rates of live birth and clinical pregnancy loss were in agreement with those of the subgroup analyses mentioned above [Supplementary Tables 10 and 11, https://links.lww.com/CM9/B923]. Then we applied logistic regression analysis to adjust potential confounders, including age, body mass index (BMI), AMH, follicle-stimulating hormone (FSH), ovarian stimulation protocols, endometrial thickness on hCG trigger day, prior spontaneous miscarriages, and antral follicle count. Consistent with the results of stratification analyses, no association was found by the logistic regression analysis between PGT-A treatment with cumulative live birth rate and clinical pregnancy loss rate among women aged <35 years [Supplementary Table 12, https://links.lww.com/CM9/B923]. Conversely, in women aged ≥35 years, PGT-A treatment was correlated with decreased cumulative clinical pregnancy loss rate (OR = 0.408, 95% CI: 0.173–0.966, P <0.05; Supplementary Table 13, https://links.lww.com/CM9/B923). This study included 1003 uRPL couples with FETs following a single oocyte retrieval. The results showed no evidence of favorable effects of PGT-A treatment on improving the cumulative live birth rate in uRPL couples regardless of maternal age (<35 years or ≥35 years). However, the risk of clinical pregnancy loss was lower after PGT-A among women with AMA (≥35 year-old) than following the conventional method. Our results showed that RPL couples had a comparable cumulative live birth rate after PGT-A. A retrospective cohort study of fertile RPL patients revealed a similar live birth rate after intent-to-treat analysis between couples with PGT-A and those under expectant management alone.[8] The ineffectiveness of PGT-A could be mainly ascribed to the high mosaicism proportion and inevitable false-positive results from trophectoderm biopsies, which leads to substantial embryo waste. Euploid embryos after PGT-A could not be obtained from about 15.8% of women aged <35 years, while the proportion was 40.5% in women aged ≥35 years in our study. In addition, trophectoderm biopsy could be a detrimental factor for embryo development and implantation potential, which was considered to be another important reason. Nonetheless, PGT-A could be beneficial in reducing the risk of pregnancy loss among AMA women rather than young women. This result could be ascribed to increased embryo aneuploidy due to AMA. Nonetheless, the decreased pregnancy loss rate did not translate into improved live birth rate because of the high proportion of no euploid embryo for transfer. The strengths of the present study are as follows. First, we performed stratified analysis according to female age and selected appropriate outcome measures, such as cumulative rates of live birth and clinical pregnancy loss following a single oocyte retrieval, which were clinically relevant. We also implemented stringent inclusion and exclusion criteria; only RPL couples undergoing PGT-A or conventional IVF for the first time were enrolled to eliminate the confounding effects. The limitations of this study were its single-center design and retrospective nature. Hence, all variables could not be controlled. In summary, there is no evidence that PGT-A improves the cumulative live birth rate in RPL couples regardless of maternal age, which could be because a large number of viable embryos were not utilized. However, PGT-A may be a favorable approach to reduce the risk of pregnancy loss in RPL women aged ≥35 years. Acknowledgments The authors would like to express their gratitude to Jingfu Yang for his invaluable assistance during data collection. The authors are appreciative of every member from the IVF and PGT laboratory for their extraordinary contributions to laboratory procedures. Funding This work was supported by grants from the National Key Research and Development Program (No. 2021YFC2700604), General Program of National Natural Science Foundation of China (No. 82171648), Shandong Provincial Key Research and Development Program (No.2021LCZX02), Youth Program of National Natural Science Foundation of China (No.82101752), Youth Program of Shandong Provincial Natural Science Foundation of China (No. ZR2021QH075), Taishan Scholars Program for Young Experts of Shandong Province (No. tsqn202312388), the Shandong Provincial Medical and Health Science and Technology Development Program (No.2019WS171) and the Shandong Provincial Natural Science Foundation (No. ZR2020MH072).