Cumulative live birth rate of in vitro fertilization cycle via progestin-primed ovarian stimulation versus gonadotropin-releasing hormone antagonist protocol in infertile women with normal ovarian reserve: an open-label, randomized controlled trial

Abstract

This study aimed to evaluate the cumulative live birth rate (cLBR) of progestin-primed ovarian stimulation (PPOS) protocol versus gonadotropin-releasing hormone antagonist (GnRH-ant) protocol for in vitro fertilization (IVF) cycle in infertile women with normal ovarian reserve (NOR). Infertile women with NOR who underwent their first IVF cycle were enrolled in an open-label randomized controlled trial. Patients were randomly assigned 1:1 to receive a freeze-all strategy with delayed embryo transfer (PPOS group, n = 174) and fresh embryo transfer first (GnRH-ant group, n = 174). The primary outcome was the cLBR per aspiration. The cLBR between the PPOS group and GnRH-ant group were comparable (55.75% vs. 52.87%, p = 0.591). A premature luteinizing hormone surge was not observed in the PPOS group, while there were six cases (3.45%) in the GnRH-ant group, but no premature ovulation in either of the groups. The pregnancy outcomes, including implantation rate, clinical pregnancy rate and miscarriage rate, were all comparable. In addition, the number of retrieved oocytes, mature oocytes and viable embryos were similar (all p > 0.05) between the two groups.

Keywords:

• Female infertility
• gonadotropin-releasing hormone antagonist
• in vitro fertilization
• ovarian stimulation
• ovarian reserve
• progestin

Trial registration:

• This trial was registered in the China Clinical Trial Registry on 6 September 2018 (ChiCTR1800018246)

Introduction

Over the past decade, novel GnRH antagonists (GnRH-ant) have been introduced as an alternative to gonadotropin-releasing hormone (GnRH) agonists for preventing premature luteinizing hormone (LH) surge during in vitro fertilization (IVF) cycles (Kalafat et al., 2022). GnRH-ant have been shown to suppress LH release by competitively binding to GnRH receptor in the pituitary immediately after the start of therapy (Copperman & Benadiva, 2013). Compared with agonist counterparts, GnRH-ant have fewer side effects (Duijkers et al., 1998; Rabinovici et al., 1992) and a lower dose requirement for a shorter duration (Zimmerman et al., 2022). Consequently, GnRH-ant have become the standard of care for reducing the incidence of premature LH surge in different clinical scenarios. A potential advantage of these drugs is that they do not produce the initial flare-up effect, thus reducing the asynchronous follicular pool recruitment (Allegra et al., 2007).

The current scenario in advanced assisted reproductive technology (ART), where embryo freezing has become an increasingly common practice, has led to the proposal of an innovative COS regimen: progestin-primed ovarian stimulation (PPOS) protocol, in which progestin is used as an alternative to GnRH analogue to prevent a premature LH surge during COS (Yu et al., 2019). Progestin gradually decreases the LH level during COS upon continuous administration, with a low incidence of LH surge and premature ovulation (Q. Chen et al., 2019; Lin et al., 2021; Yu et al., 2019). When used in combination with delayed embryo transfer (ET), the PPOS protocol results in a comparable yield of retrieved oocytes, viable embryos, and pregnancy outcome to that of the short protocol and the GnRH-ant protocol in infertile women with normal ovarian reserve (NOR) (Beguería et al., 2019; Q. Chen et al., 2017; Dong et al., 2017; Kuang et al., 2015), and PPOS has the convenience of an oral agent, lower cost and no injection pain. Wang et al. (2016) showed that medroxyprogesterone acetate (MPA) could inhibit premature luteinization in women undergoing IVF and lead to a lower incidence of the ovarian hyperstimulation syndrome (OHSS) in women with polycystic ovarian syndrome (PCOS). A review by Massin (2017) showed that PPOS effectively suppresses premature LH surge and reduces the incidence of OHSS when combined with a freeze-all strategy and GnRH-agonist trigger.

In patients with low prognosis diagnosed according to the POSEIDON criteria, the pregnancy rates were similar between the PPOS and GnRH-ant protocols, and patients who received the GnRH-ant protocols even showed a shorter timer to achieve a live birth (Du et al., 2021). Zhu et al. (2021) and Q. Chen et al. (2019) both showed that the PPOS and GnRH-ant protocols were similar regarding oocyte yield and pregnancy rates but that the LH rise could be more easily controlled with the PPOS protocol. On the other hand, PPOS could decrease the incidence of OHSS without adversely affecting clinical outcomes in patients with PCOS (Xiao et al., 2019). A meta-analysis suggested that the PPOS protocol was suitable for patients with different ovarian reserves, including NOR (Guan et al., 2021). However, the previous meta-analysis was mainly based on observational studies and its conclusions need to be validated in a study with a robust design. Therefore, we established this randomized controlled study to evaluate the cumulative live birth rate (cLBR) in infertile women with NOR who received the PPOS and GnRH-ant protocols.

Materials and methods

Study design and population

This open-label randomized controlled trial was conducted at the reproductive centre of Shanghai East hospital between September 2018 and December 2019. Infertile women aged 20–40 years who were undergoing their first cycle of IVF and had a spontaneous menstrual cycle (25–35 days), serum anti-Müllerian hormone (AMH) levels > 1.1 ng/ml, bilateral antral follicular count (AFC) 8–20, basal follicle-stimulating hormone (FSH) < 10 mIU/mL were enrolled. Women who had basal oestradiol levels > 80 pg/mL, a recurrent miscarriage with two or more spontaneous abortions, requiring preimplantation genetic testing, or any contraindications for COS or were diagnosed with endometriosis (grade 3 or higher), PCOS, hydrosalpinx, genital tract tumours, abnormal karyotype, abnormal uterine anatomical structure were excluded.

This study was performed in accordance with the Declaration of Helsinki for medical research and Good Clinical Practice guidelines and was approved by the Ethics Committee of the Institutional Review Board (IRB) of our hospital (number: 2018-17). This trial was registered in the China Clinical Trial Registry on 6 September 2018 (ChiCTR-1800018246). Written informed consent was obtained from all enrolled patients who underwent their ovarian stimulation cycle.

Randomization

A computer-generated list was used for randomization. Women who met the eligibility criteria were randomly assigned to either the PPOS or the GnRH-ant group at a ratio of 1:1. Treatment allocation list was generated by the investigators and then given to the researchers in sealed numbered envelopes. A physician prescribed the drugs specific to the code, while an experienced nurse instructed the patients on the drug dosage regimen. The physicians, nurses, and embryologists were aware of the allocation.

Intervention

PPOS protocol

Transvaginal ultrasonography was performed on day 2 or 3 of menstruation, and basal serum hormone (FSH, LH, oestradiol [E2] and progesterone) levels were monitored to exclude any patients with cysts or elevated basal oestradiol levels. Subsequently, patients were given medroxyprogesterone acetate (MPA, Shanghai Xinyi Pharmaceutical Co., China) at a dose of 10 mg/day orally in combination with intramuscular injection urinary human menopausal gonadotropin (hMG, Anhui Fengyuan Pharmaceutical Co., China) at a dose of 150–225 IU/day from day 2 or 3 of menstruation. The hMG dosage was adjusted according to the serum E2 levels and follicle sizes, while the dose of MPA remained constant. When three or more dominant follicles reached 18 mm in diameter, the final oocyte maturation was triggered by administering subcutaneous GnRH agonist (Triptorelin, Decapeptyl®; Ferring Pharmaceuticals, Germany) at a dose of 0.1 mg and/or urinary human chorionic gonadotropin (hCG 1000 IU, Lizhu Pharmaceutical Trading Co., China) intramuscular injection. Transvaginal ultrasound-guided oocyte retrieval was performed 36 h later following the trigger.

GnRH antagonist protocol

The flexible GnRH-ant protocol was used in this study. COS was performed by administering hMG at a dose of 150–225 IU/day from day 2 or 3 of menstruation until the trigger day. GnRH-ant (Cetrotide, 0.25 mg, Merck-Serono) was administered at a dose of 0.25 mg/day when serum LH levels increased (≥10 IU/L) or when the diameter of the leading follicles was found to be > 14 mm. The dosage of hMG was adjusted according to the patient’s characteristics, serum LH level, and follicular response. The final stage of oocyte maturation was also triggered with Triptorelin (0.1 mg) and/or hCG injections, and oocyte retrieval was performed about 36 h following the trigger. Patients with high response did not receive hCG trigger, and fresh ET was also cancelled.

Oocyte collection and embryo culture

Oocyte retrieval was performed as per routine protocol. Briefly, all follicles with a diameter of >10 mm were aspirated. Standard insemination or ICSI was performed, as appropriate, within 6 h of retrieval. On day 3, the number of embryos, the regularity of blastomeres, and the degree of embryonic fragmentation were examined. All the best-quality cleavage-stage embryos obtained from the PPOS group were frozen, while extended culture was carried out for embryos of other grades, followed by cryopreservation. One or two best-grade embryos were transferred to the GnRH-ant group if the patient’s condition permitted fresh ET. At the blastocyst stage, embryo quality assessment was performed based on the Gardner and Schoolcraft grading system (Copperman & Benadiva, 2013). Only blastocysts with good morphology were frozen on day 5 or day 6.

Endometrium preparation and frozen ET

In the case of the GnRH-ant group, the freeze-all strategy was performed only in (i) patients who were at high risk of OHSS or (ii) patients who exhibited elevated serum progesterone levels (>1.5 ng/ml) on trigger day or (iii) patients who specifically opted for the procedure.

Frozen ET was conducted at least after the second menstrual cycle following oocyte pick-up. The selection of endometrial priming protocol (natural cycle, mild stimulation or hormone replacement therapy) for frozen ET was based on a combination of routine laboratory results obtained in the present study and a previously described approach (Kuang et al., 2015). Briefly, natural cycle and mild stimulation were the first choice for the endometrial preparation, and hCG was used to induce the final ovulation. Patients were given 20 mg of oral dydrogesterone (10 mg/capsule; Duffon, Shanghai Abbott Pharmaceutical Co., Ltd.) and 0.2 g of soft progesterone vaginal capsules (0.1 g/capsule; Angitan, Besins Manufacturing Belgium) twice daily for luteal support after the day of theoretical day of ovulation (Kuang et al., 2015). The timing of frozen ET was performed three or five days later after ovulation. Hormone replacement therapy was recommended for patients with thin endometrium. For the cycle without ovulation, 4 mg of oral Femoston progesterone capsule (2 mg/capsule; Solvay Pharmaceutical Co., Ltd.) and 0.2 g of soft progesterone vaginal capsules were administered twice daily after endometrial transform. Cleavage embryo and blastocyst transfer was performed three and five days after exogenous progesterone exposure respectively. Up to two embryos were transferred. When pregnancy was achieved, exogenous progesterone supplementation was continued until 10 weeks of gestation.

Outcomes

The definitions in this study were according to the consensus definition in 2017 (Zegers-Hochschild et al., 2017). The primary outcome was cLBR, defined as the number of deliveries with at least one live birth per oocyte aspiration. A complete ovum pickup cycle was defined as a cycle in which a delivery with a live birth occurred or all embryos were used, whichever occurred first. The secondary outcome measures included the number of oocytes retrieved, incidence of premature LH surge, duration and dosage of hMG treatment, number of mature oocytes and viable embryos, and implantation and pregnancy rates. Safety measures included the incidence of serious adverse events, such as moderate/severe OHSS, bleeding, or infection following oocyte retrieval. A high ovarian response was defined as >15 retrieved oocytes, while a poor response was defined as < 5 retrieved oocytes. Therefore, a normal response was defined as 5–15 retrieved oocytes. Positive pregnancy was defined as positive by an elevated serum β-hCG level and there was no visible gestational sac on ultrasound two weeks following ET. Clinical pregnancy was defined as at least one gestational sac being observed on ultrasound after ET. A premature LH surge was defined as a serum LH level of >15 mIU/mL and a progesterone level of ≥ 1.0 ng/mL. OHSS was classified into three grades based on its severity (mild, moderate, and severe) as per 2016 guidelines (Practice Committee of the American Society for Reproductive Medicine, 2016).

Statistical analysis

According to the cLBR estimated at 35% of the GnRH-ant protocol in previous literature (Toftager et al., 2017) and actual clinical data of the young patients with normal ovarian reserve in our centre, a priori statistical power analysis (PASS, version 11, NCSS, WI, USA) indicated that a sample size of 167 patients in each group was required to achieve at least 80% power at an α level of 0.05 with an cLBR of 50%. Given the possibility of a 5% dropout rate, we designed the study to include a total of 174 patients in each group.

The primary outcome was analysed according to the intention-to-treat principle. All data were analysed using the SPSS, version 25.0 (IBM Corp., Armonk, N.Y., USA). The normality of data distribution was assessed by the Shapiro-Wilk test. Continuous data were expressed as mean ± standard deviation (SD) for the normal distribution or non-normal distribution data were expressed as median (interquartile range). Categorical data were expressed as n (%) and compared by the Chi-square test. The results were compared between the two groups using unpaired Student’s t-test, nonparametric Mann–Whitney U-test, Pearson’s χ² test. A two-sided p value of < 0.05 was considered statistically significant.

Results

A total of 348 patients were enrolled: 174 patients allocated to each group. In the PPOS group, oocyte was not obtained from one patient. However, no such cases were observed in the GnRH-ant group. Four patients in the PPOS group and two in the GnRH-ant group had no viable embryos. By the end of this study period, in the PPOS group, out of 170 patients who had viable embryos, 142 patients had completed 200 frozen ET cycles. In the GnRH-ant group, out of 172 patients with viable embryos, 80 patients underwent 80 fresh ET cycles, while 82 patients completed 115 frozen ET cycles. 17 patients underwent both fresh ETs and frozen ET cycles. In addition, 32 patients in the PPOS group and 29 patients in the GnRH-ant group did not undergo ET. A total of 63 patients did not undergo fresh ET due to one of the following reasons: patient choice, high serum progesterone levels on the day of hCG administration, and risk of OHSS. Finally, 97 and 92 live births were recorded in the PPOS and GnRH-ant groups, respectively. The flowchart of patients’ screening and selection process in this study is shown in Figure 1.

Figure 1. Flowchart of the screening and selection process in this prospective randomized controlled trial.

Patients’ age, body mass index (BMI), duration of infertility, primary infertility factors, cause of infertility, AFC, AMH level, and baseline hormonal profile between PPOS and GnRH-ant groups were comparable (all p > 0.05) (Table 1).

Table 1. Baseline characteristics.

Characteristics	PPOS with MPA (n = 174)	GnRH antagonist (n = 174)	p Value
Age (years), mean ± SD	32.12 ± 3.69	32.11 ± 4.1	0.989
BMI (kg/m^2), mean ± SD	21.85 ± 3.2	21.75 ± 3.2	0.785
Duration of infertility (years), median (interquartile range)	3 (1.5)	3 (1.5)	0.095
Primary infertility factor, n (%)	123 (70.69)	114 (65.52)	0.301
Cause of infertility, n (%)
Tube factor	93 (53.45)	90 (51.72)	0.830
Male factor	41 (23.56)	38 (21.84)	0.798
Combined factors	35 (20.11)	40 (22.99)	0.602
Unexplained factors	5 (2.88)	6 (3.45)	0.759
AFC (n), mean ± SD	9.92 ± 3.89	9.98 ± 4.22	0.885
AMH (ng/mL), mean ± SD	3.41 ± 1.73	3.63 ± 2.48	0.560
Baseline hormonal profile, mean ± SD
bFSH (mIU/mL)	7.17 ± 1.6	7.31 ± 1.76	0.440
bLH (mIU/mL)	4.22 ± 2.17	4.23 ± 2.16	0.973
bE2 (pg/mL)	40.52 ± 12.14	39.99 ± 13.5	0.701
bP (ng/mL)	0.53 ± 0.48	0.53 ± 0.52	0.887

Abbreviations: BMI: body mass index; GnRH: gonadotropin releasing hormone; MPA: medroxyprogesterone acetate; PPOS: progesterone-primed ovarian stimulation. AFC: antral follicle count; AMH: anti-Mullerian hormone.

Although the number of oocyte retrieved in the PPOS group was slightly higher compared with that of the GnRH-ant group (9.88 ± 5.31 vs. 9.14 ± 5.12; p = 0.185), the difference was not statistically significant. No significant differences were found in the oocyte retrieval rate (82.69% vs. 83.34%; p = 0.612), number of metaphase stage II oocytes, (8.13 ± 4.66 vs. 7.49 ± 4.23; p = 0.182), number of best grade embryos obtained on day 3 (4.04 ± 2.60 vs. 3.80 ± 2.25; p = 0.385), viable embryos (4.60 ± 2.58 vs. 4.44 ± 2.03; p = 0.518) and proportion of viable embryos per oocyte (46.95% vs. 48.59%; p = 0.348) between the two groups. Notably, the time interval between the trigger and oocyte pickup (36.28 ± 0.66 vs. 35.71 ± 1.74; p = 0.001) was found to be significantly different between PPOS and GnRH-ant groups. Compared with the PPOS group, the stimulation duration (9.03 ± 1.56 vs. 8.64 ± 1.75, p = 0.029) was found to be significantly lower in GnRH-ant group. However, no significant differences were observed in hMG dosage (1909.05 ± 421.77 vs. 1828.88 ± 503.77; p = 0.108) between the two groups.

The mean LH level on the trigger day was significantly lower in PPOS group than GnRH-ant group (2.30 ± 1.78 vs. 3.66 ± 3.52, p < 0.01), accompanied with a lower progesterone level corresponding (0.74 ± 0.41 vs. 0.99 ± 0.63, p < 0.01). However, the progesterone value on the day after trigger was similar between the two groups. No incidence of premature LH surge was found in PPOS group, whereas six cases had premature LH surge in GnRH ant group (ranged from 15.05–22.70 mIU/ml) on the trigger day but all of them achieved viable embryos (ranged from 2 to 11). Three cases in the PPOS group had the LH level >10 mIU/ml and 10 cases in the GnRH-ant group.

There were no cases of premature ovulation and severe OHSS in both groups (Table 2).

Table 2. Outcomes of embryology and stimulate ovulation.

Parameters	PPOS with MPA (n = 174)	GnRH antagonist (n = 174)	p Value
Duration of stimulation (days), mean ± SD	9.03 ± 1.56	8.64 ± 1.75	0.029*
Total hMG dosage (IU), mean ± SD	1909.05 ± 421.77	1828.88 ± 503.77	0.108
LH-value on trigger day (mIU/mL), mean ± SD	2.30 ± 1.78	3.66 ± 3.52	0.000*
p Value on trigger day (ng/mL), mean ± SD	0.74 ± 0.41	0.99 ± 0.63	0.000*
p Value on the day after trigger (ng/mL), mean ± SD	3.80 ± 1.91	4.17 ± 2.39	0.110
Trigger drugs			0.081
GnRH-a	16.67 (29)	9.77 (17)
GnRH-a and hCG	83.33 (145)	90.23 (157)
Oocytes retrieved, mean ± SD	9.88 ± 5.31	9.14 ± 5.12	0.185
Rate of oocyte retrieval (%)	82.69 (1706/2063)	83.34 (1591/1909)	0.612
Fertilization methods			0.798
IVF(%)	76.44 (133)	78.16(136)
ICSI(%)	23.56 (41)	21.84 (38)
Metaphase stage II oocytes (n), mean ± SD	8.13 ± 4.66	7.49 ± 4.23	0.182
Best grade embryos on day 3 (n), mean ± SD	4.04 ± 2.60	3.80 ± 2.25	0.385
Viable embryos (n), mean ± SD	4.60 ± 2.58	4.44 ± 2.03	0.518
Proportion of viable embryos per oocytes (%)	46.95 (801/1706)	48.59 (773/1591)	0.348
Time interval between trigger and OPU (h), mean ± SD	36.28 ± 0.66	35.71 ± 1.74	0.001*
Premature ovulation (n)	0	0	–
Incidences of LH surge (n)	0	6	–
Severe OHSS (n)	0	0	–

Abbreviations: hMG: human menopausal gonadotropin; GnRH: gonadotropin releasing hormone; hCG: human chorionic gonadotropin; IVF: in vitro fertilization; ICSI: intracytoplasmic sperm injection; MPA: medroxyprogesterone acetate; OHSS: ovarian hyperstimulation syndrome; LH: luteinizing hormone. *p Value <0.05 is considered as significant.

Table 3 shows the pregnancy outcomes following frozen and fresh ET. The cLBR (55.75% vs. 52.87%, p = 0.591) were comparable between PPOS group and GnRH-ant group.

Table 3. Pregnancy outcomes following frozen and fresh embryo transfer.

Outcomes	PPOS with MPA (n = 174)	GnRH antagonist (n = 174)	p Value
cLBR, %	55.75	52.87	0.591
Fresh ET cycles (n)	–	80	–
Frozen ET cycles (n)			–
No. of frozen ET1 cycles	98	55
No. of frozen ET2 cycles	30	21
No. of frozen ET3 cycles	14	6
Endometrial thickness on ET day (mm), mean ± SD
Fresh ET cycles	–	10.68 ± 1.98	–
Frozen ET cycles	10.75 ± 2.05	10.42 ± 2.08	0.169
The number of transferred embryos (%) (n/total)			0.570
Single	13.50 (27/200)	15.90 (31/195)
Two	86.50 (173/200)	84.10 (164/195)
Stages at which embryos were transferred (%) (n/total)			0.032*
Cleavage stage	90.88 (339/373)	94.99 (341/359)
Blastocyst stage	9.12 (34/373)	5.01 (18/359)
Pregnancy outcomes
Implantation rate (%)	37.27	36.77	0.939
Positive pregnancy rate (%)(n/total)	64 (128/200)	63.59 (124/195)	0.932
Clinical pregnancy rate (%) (n/total)	57 (114/200)	55.89 (109/195)	0.825
Miscarriage rate (%)(n/total)	12.28 (14/114)	13.76 (15/109)	0.742
Ectopic pregnancy rate (%)(n/total)	1.75 (2/114)	0.92 (1/109)	1.000
LBR in the different ovarian response			0.199
Low ovarian response (%)(n)	36 (9/25)	30 (9/30)
High ovarian response (%)(n)	58.62 (17/29)	52.94 (9/17)
Normal ovarian response (%)(n)	59.17(71/120)	58.27(74/127)
Twin pregnancy rate (%)(n)	22.81 (26/114)	22.94 (25/109)	0.982
Gestation (weeks), mean ± SD	38.25 ± 1.45	38.52 ± 1.68	0.762
Newborn birth weight (g), mean ± SD
Singleton	3330.27 ± 487.52	3347.64 ± 393.53	0.837
Twin	2404.63 ± 548.15	2464 ± 331.02	0.527

Abbreviations: cLBR: cumulative live birth rate; ET: embryo transfer; GnRH: gonadotropin releasing hormone; MPA: medroxyprogesterone acetate; PPOS: progesterone-primed ovarian stimulation; *p Value <0.05 is considered as significant.

At the end of the study period, 142 patients in the PPOS group had completed 200 frozen ET cycles, while those in the GnRH-ant group completed 195 cycles (including 80 fresh embryo cycles and 115 frozen ET cycles). All the transferred embryos were cleavage-stage embryos for fresh cycles, whereas both cleavage-stage embryos and blastocysts were transferred for frozen ET cycles. Notably, a statistically significant difference was observed in the proportion of stages at which embryos were transferred (p = 0.032) between the two groups. The number of transferred embryos (p = 0.570), implantation rate (37.27% vs. 36.77%; p = 0.939), positive pregnancy rate (64% vs. 63.59%; p = 0.932), and clinical pregnancy rate (55.75% vs. 55.89%; p = 0.825) were comparable between the two groups. Furthermore, there were no significant differences in the miscarriage rate (12.28% vs. 13.76%, p = 0.742). There were two cases (n = 2) of ectopic pregnancy in the PPOS group in contrast to one case (n = 1) in the GnRH-ant group. There were no statistically significant differences in LBR in different ovarian responses (p = 0.199), twin pregnancy rate (22.81% vs. 22.94%, p = 0.982), and the birth weights of singleton (3330.27 ± 487.52 vs. 3347.64 ± 393.53; p = 0.837) and twin (2404.63 ± 548.15 vs. 2464 ± 331.02; p = 0.527) neonates between the two groups.

Discussion

Main findings

The present data demonstrated that IVF via PPOS protocol have a similar cLBR compared to the GnRH-ant protocol. PPOS and GnRH-ant regimens yielded a comparable number of retrieved oocytes and embryos and similar pregnancy outcomes in infertile women with NOR undergoing IVF. Furthermore, the ability of oral progestin to prevent premature ovulation was also found to be similar to that of the GnRH-ant.

Strengths

To the best of our knowledge, this is the largest open-label randomized trial to investigate the efficiency of the PPOS and GnRH-ant protocol in women with NOR undergoing IVF.

Weakness

This study was limited by its single-centre design, the shorter duration of follow-up. Multicentre prospective analysis for more generalizable findings should be conducted to assess cLBR in patients of different ages and compare cLBR with a longer duration of follow-up following the transfer of embryos from different stages. In addition, there were about 17.2% patients who did not have ET in both groups, one possible explanation was that there were not enough embryos to transfer or cycle cancelled was repeatedly due to a thin endometrium (< 7 mm).

Comparison with other studies

Several previous studies, including RCTs, reported similar cLBR in poor responders (Q. Chen et al., 2019; Du et al., 2021; Guan et al., 2021; Wang et al., 2018) and women with PCOS (Zhu et al., 2021) stimulated with PPOS compared with the GnRH-ant protocol. Consistent with these studies, our finding showed comparable cLBR in PPOS and GnRH-ant groups. On the other hand, H. Chen et al. (2022) reported that the GnRH-ant protocol yielded higher cLBRs and shorter time to live birth than the PPOS protocol in unselected women undergoing IVF. However, that study pooled together vastly different patients in terms of age, causes of infertility and prognosis, while patients older than 40 were not included in this study. Zhang et al. (2021) also reported a better cLBR with the GnRH-ant protocol than the PPOS protocol in patients diagnosed according to the POSEIDON criteria. However, the patients closer to the ones included in the present study are POSEIDON groups 1 and 2, and there were no significant differences between the GnRH-ant and PPOS protocols in POSEIDON groups 1 and 2, supporting the present study.

Though an increase in LH on the trigger day was accompanied by a corresponding increase in progesterone levels was observed in GnRH-ant protocol, there was no premature ovulation, and the results showed no significant differences in oocytes and embryos outcomes between the two groups. In addition, a shorter stimulation duration and a shorter interval between trigger and oocyte pickup were required in the GnRH-ant group, which was associated with the different LH suppression mechanism between PPOS and GnRH-ant protocols. Studies in animal models have shown that progesterone-mediated suppression of LH surge is mediated by hypothalamic dynorphin and gamma-aminobutyric acid (GABA) receptor signalling and kisspeptin (He et al., 2017; Liu et al., 2020), whereby serum LH levels decline slowly and steadily during the PPOS treatment. In contrast, GnRH-ant-mediated pituitary secretion occurs rapidly due to the direct and competitive blockade of GnRH receptors in the late follicular phase. This competitive nature of pituitary suppression may be one of the reasons for the occasional escape in the GnRH-ant protocol.

Although the cut-off value for LH surge differed among different centres, pregnancy outcome rates remained constant in frozen ET cycles. In this study, an increase in LH level of ≥ 10 mIU/ml was observed in 3 patients of the PPOS group and 10 patients of the GnRH-ant group. Interestingly, all 3 patients in the PPOS group and four patients in the GnRH group who underwent fresh ET produced live births. And despite the premature LH surge observed in six patients of the GnRH group, oocytes and viable embryos were obtained due to a flexible advanced arrangement of oocyte retrieval in our experience (Lu et al., 2018; Shen et al., 2020). Based on this, it necessitates the need for studies to investigate the cut-off level for premature LH surge. Another interesting observation of this study is that low LH levels still did not affect the reproductive outcome except for endogenous increased in LH. However, a retrospective cohort study by Luo et al. (2021) on GnRH-ant protocol reported that low LH levels (< 4 IU/L) led to a decrease in live birth rates in fresh ET cycles but not in the freeze-all cycles. Furthermore, they found that the cLBRs significantly decreased in those with low LH (63.1% vs. 68.3%, p = 0.034) compared to those with higher LH. These findings raise the question of whether serum LH levels can still be used as an effective biomarker during COS. Therefore, monitoring LH levels throughout COS may help physicians in making decisions.

A potential advantage associated with the freeze-all regimen is that it may provide a more physiological hormonal and endometrial environment for embryo (Wei et al., 2019). On the other hand, for the GnRH-ant protocol group, fresh cycles were performed first, and frozen blastocyst transfer was performed after the failure of fresh transfer. Fresh cycle transfers were almost all performed with embryos at the cleavage-stage to prevent transfer cycle cancellation owing to no blastocysts for transfer. Furthermore, the first, second, and third ET cycles of both groups were analysed, and the results showed that cLBR were similar between the two treatment groups. Consistent with a recent study, cLBR obtained was comparable by initial ET followed by frozen embryo transfer both in the PPOS group and GnRH-ant group in oocyte donation (Giles et al., 2021; Yildiz et al., 2019). These findings are in contrast to a study carried out by Zacà et al. (2018), which reported the occurrence of superior cLBR outcomes using frozen blastocyst transfer in the frozen ET cycle. This could possibly be attributed to variability in the strategy. Our results, therefore, indicate that embryos derived from the PPOS group may possess an identical developmental potential as those originating from the GnRH-ant group.

The GnRH-ant protocol is preferable for women with a good prognosis who would expect a shorter time to pregnancy. Under this protocol, frequent monitoring is necessary, and the protocol has an unavoidable risk of late OHSS induced by pregnancy (La Marca & Capuzzo, 2019; Namavar Jahromi et al., 2018). At the same time, PPOS is a new, simplified protocol for ovarian stimulation that is appropriate for the freeze-only strategy. Undoubtedly, PPOS offers several advantages: the need for fewer injections, and being patient-friendly. However, it also has the disadvantages of delayed ET and an extended treatment time to achieve pregnancy. Currently, PPOS is considered only for patients who require all embryos to be frozen, such as patients with cancer and those who opt for fertility preservation.

Implications for clinical practice

The results of this study demonstrates the efficacy and safety of PPOS, indicating PPOS will play a promising role in the freeze-only strategy via its advantages of patient friendliness and simplified administration.

Implications for future research

It is critical and urgent to conduct a multi-centre RCT that will involve a larger patient cohort, which will compare GnRH-ant protocol and PPOS protocol in terms of time-cost-effectiveness per live birth, individuals with PCOS and DOR and safety of offspring.

Ethical approval

This work has been carried out in accordance with the Declaration of Helsinki (2000) of the World Medical Association. This study was approved by the Ethics Committee of Shanghai East Hospital (Institutional Review Board number: 2018-17).

Consent form

All patients provided written informed consent to participate in this study.

Author contributions

Lihua Sun, Qiuju Chen, and Hongjuan Ye were the chief investigators who completed the entire study, including procedures, conception, design, and completion. Hongjuan Ye also drafted the manuscript. Liya Shi and Xinxin Quan analysed the data. Min Hou and Zhao Yu were responsible for data collection. Huilan Ma contributed to the follow-up. Songguo Xue was in charge of embryos. All authors approved the final manuscript.

Acknowledgements

We would like to thank all patients and the entire staff of the reproductive centre of Shanghai East Hospital for their support in this trial.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

All data generated or analysed during this study are included in this article.

Additional information

Funding

This study was funded by the Natural Science Foundation of Shanghai (No. 18411964000) and the National Natural Science Foundation of China (No. 82101734).