Feasibility and efficacy of random-start progestin-primed ovarian stimulation: a multicenter pilot randomized controlled trial in oocyte donation.

The participant flow chart in the study is presented in Fig. 1 . Among the 110 egg donors randomized in the trial, 23 were included in the EFP, 22 in the MFP, 21 in the LFP, 21 in the OP and 23 in the LP group. Finally, in the intention-to-treat analysis, 22, 20, 19, 18 and 23 donors were assigned respectively to the EFP, MFP, LFP, OP and LP groups. The per-protocol analysis was performed on 21, 25, 18, 15 and 23 egg donors, respectively, in the EFP, MFP, LFP, OP and LP groups. Baseline characteristics of oocytes donors were similar between groups in the per protocol analysis, with a median age of 33 [29.2;35] years, BMI of 23.4 [21.1;26.3] kg/m 2 , Anti-Mullerian Hormone (AMH) of 3 [2.2;4.1] ng/ml and an antral follicle count (AFC) of 21 [15.2;28] (Table 1 ). Similarly, baseline characteristics were identical between groups in intention-to-treat analysis (Supplementary Table 1 ). Table 1 Demographic data per group in the per-protocol population EFP ( N =21) MFP ( N =25) LFP ( N =18) OP ( N =15) LP ( N =23) Total ( N =102) Age, yrs  Median(Q1-Q3) 33.0 (29.0, 36.0) 33.0 (31.0, 35.0) 33.0 (31.0, 34.0) 33.0 (27.5, 34.0) 32.0 (26.0, 35.5) 33.0 (29.2, 35.0)  Mean(SD) 31.9 (4.8) 32.8 (3.1) 32.6 (2.7) 30.9 (4.7) 30.4 (5.3) 31.7 (4.3)  Min-Max 20.0 - 37.0 25.0 - 37.0 27.0 - 37.0 23.0 - 37.0 21.0 - 37.0 20.0 - 37.0 Weight, Kg  Median(Q1-Q3) 65.0 (58.0, 75.0) 67.0 (62.0, 75.0) 67.0 (60.5, 72.5) 63.0 (56.5, 71.0) 58.0 (55.0, 65.0) 64.5 (57.2, 72.8)  Mean(SD) 66.0 (9.8) 68.0 (11.4) 66.9 (10.5) 66.2 (12.9) 61.8 (10.0) 65.7 (10.9)  Min-Max 50.0 - 84.0 44.0 - 90.0 50.0 - 90.0 50.0 - 90.0 50.0 - 85.0 44.0 - 90.0 BMI, Kg/m²  Median(Q1-Q3) 23.4 (20.8, 25.3) 25.5 (22.3, 27.2) 23.8 (21.1, 26.9) 23.4 (21.9, 25.6) 21.6 (20.2, 24.5) 23.4 (21.1, 26.3)  Mean(SD) 23.5 (3.1) 25.2 (3.8) 24.4 (3.9) 24.1 (3.3) 22.7 (3.4) 24.0 (3.6)  Min-Max 19.0 - 29.1 19.3 - 31.9 19.6 - 31.5 19.1 - 31.1 19.0 - 31.6 19.0 - 31.9 AMH (ng/ml)  NA 2 0 0 1 0 3  Median(Q1-Q3) 3.0 (2.2, 3.6) 3.0 (2.4, 4.1) 3.5 (2.2, 4.6) 3.2 (2.9, 3.9) 2.4 (2.0, 4.0) 3.0 (2.2, 4.1)  Mean(SD) 2.9 (0.9) 3.2 (1.5) 3.7 (1.9) 3.4 (1.1) 3.0 (1.6) 3.2 (1.5)  Min-Max 1.1 - 4.3 0.8 - 6.6 1.5 - 8.8 1.2 - 5.6 0.3 - 7.4 0.3 - 8.8 Antral follicle count  Median(Q1-Q3) 18.0 (15.0, 23.0) 21.0 (16.0, 27.0) 21.0 (16.2, 26.8) 24.0 (20.0, 31.5) 24.0 (15.0, 28.5) 21.0 (15.2, 28.0)  Mean(SD) 20.5 (7.6) 23.0 (10.3) 22.8 (8.9) 26.3 (10.7) 23.1 (9.6) 23.0 (9.4)  Min-Max 10.0 - 39.0 8.0 - 50.0 10.0 - 44.0 6.0 - 50.0 11.0 - 46.0 6.0 - 50.0 EFP Early follicular phase, MFP Mid follicular phase, LFP Late Follicular phase, OP Ovulatory phase, LP Luteal phase, NA Not available Demographic data per group in the per-protocol population EFP Early follicular phase, MFP Mid follicular phase, LFP Late Follicular phase, OP Ovulatory phase, LP Luteal phase, NA Not available The number of retrieved oocytes did not differ between the five groups in the per-protocol analysis with a median of 11 in the EFP and MFP groups, 15 in the LFP group, 20 in the OP group and 10 in the LP group ( p  = 0.314). Similarly, the median number of mature oocytes was equivalent between per protocol groups (Table 2 ). Table 2 Comparison of number of oocytes retrieved and secondary outcomes among the treatment groups in the perprotocol population EFP ( N =21) MFP ( N =25) LFP ( N =18) OP ( N =15) LP ( N =23) Total ( N =102) p value Number of oocytes retrieved 0.314  Median(Q1-Q3) 11.0 (9.0,17.0) 11.0 (9.0,19.0) 15.0 20.8) 20.0 (8.5, 24.5) 10.0 (6.5,16.0) 11.5 (7.0,19.0)  Mean(SD) 13.1 (7.4) 13.2 (8.0) 15.8 (9.8) 18.7 (11.7) 12.2 (8.6) 14.2 (9.1)  Min-Max 2.0 - 35.0 2.0 - 30.0 4.0 - 38.0 1.0 - 41.0 1.0 - 39.0 1.0 - 41.0 Number of mature oocytes 0.335  Median(Q1-Q3) 10.0 (7.0, 14.0) 9.0 (6.0, 14.0) 13.0 (6.0, 17.0) 16.0 (8.5, 21.5) 10.0 (4.5, 15.0)  Mean(SD) 11.3 (6.3) 10.4 (6.1) 12.6 (7.5 15.9 (10.1) 10.7 (7.9)  Min-Max 2.0 - 27.0 2.0 - 24.0 3.0 - 31.0 0.0 - 39.0 0.0 - 36.0 Mild Ovarian hyperstimulation syndrome 0.586  No 19 (95.0%) 23 (92.0%) 16 (94.1%) 15 (100.0%) 23 (100.0%)  Yes 1 (5.0%) 2 (8.0%) 1 (5.9%) 0 (0.0%) 0 (0.0%) Premature LH surge 0.540  No 21 (100.0%) 24 (96.0%) 18 (100.0%) 15 (100.0%) 23 (100.0%)  Yes 0 (0.0%) 1 (4.0%) 0 (0.0%) 0 (0.0%) 0 (0.0%) Additional dose of rFSH 0.005  No 0 (0.0%) 1 (4.0%) 4 (22.2%) 0 (0.0%) 0 (0.0%)  Yes 21 (100.0%) 24 (96.0%) 14 (77.8%) 15 (100.0%) 23 (100.0%) Total number of rFSH units 0.491  Median(Q1-Q3) 600.0 (400.0,900.0) 400.0 (300.0, 900.0) 300.0 (200.0, 937.5) 600.0 (387.5,1137.5) 900.0 (425.0,1200.0)  Mean(SD) 656.0 (365.0) 649.5 (551.0) 580.6 (625.7) 813.3 (550.1) 823.9 (462.9)  Min-Max 200.0 - 1500.0 0.0 - 2250.0 0.0 - 2100.0 200.0 - 2000.0 0.0 - 1500.0 Duration of ovarian stimulation (days) 0.345  NA 0 0 0 0 1 1  Median(Q1-Q3) 10.0 (9.0, 11.0) 10.0 (9.0, 11.0) 9.0 (8.0, 10.8) 10.0 (9.0, 11.5) 11.0 (9.2, 11.8)  Mean(SD) 9.8 (1.2) 10.1 (2.2) 9.6 (2.2) 10.6 (2.0) 10.6 (1.4)  Min-Max 8.0 - 12.0 7.0 - 16.0 7.0 - 14.0 8.0-15.0 8.0 - 13.0 Total number of injections 0.101  Median(Q1-Q3) 5.0 (4.0, 6.0) 4.0 (4.0, 5.0) 4.0 (3.0, 5.0) 5.0 (4.0, 6.5) 6.0 (4.5, 7.0)  Mean(SD) 5.0 (1.4) 4.7 (1.6) 4.4 (2.2) 5.5 (1.8) 5.7 (1.4)  Min-Max 3.0 - 8.0 2.0 - 8.0 2.0 - 9.0 3.0 - 10.0 3.0 - 8.0 Overall treatment satisfaction 0.645  NA 1 4 1 1 1  Median(Q1-Q3) 8.5 (8.0, 9.2) 9.0 (8.0, 10.0) 9.0 (8.0, 9.0) 9.0 (8.2, 9.8) 8.5 (7.2, 10.0)  Mean(SD) 8.3 (1.5) 8.9 (1.3) 8.6 (1.0) 8.9 (1.0) 8.4 (1.6)  Min-Max 4.0 - 10.0 6.0 - 10.0 7.0 - 10.0 7.0 - 10.0 5.0 - 10.0 EFP Early follicular phase, MFP Mid follicular phase, LFP Late Follicular phase, OP Ovulatory phase, LP Luteal phase. Kruskal-Wallis test, ANOVA, Pearson’s Chi-square test, Fisher’s exact testwere used. LH surge: defined as ≥ 10 UI/ml Minimal or mild OHSS Comparison of number of oocytes retrieved and secondary outcomes among the treatment groups in the perprotocol population EFP Early follicular phase, MFP Mid follicular phase, LFP Late Follicular phase, OP Ovulatory phase, LP Luteal phase. Kruskal-Wallis test, ANOVA, Pearson’s Chi-square test, Fisher’s exact testwere used. LH surge: defined as ≥ 10 UI/ml Minimal or mild OHSS In intention-to-treat, results were similar between the five groups, concerning the total number of retrieved oocytes ( p  = 0.473) and the number of mature oocytes ( p  = 0.592) (Supplementary Table 2 ). Concerning the difference between the total number of oocytes retrieved in the « Late Follicular Phase » and « Ovulatory Phase » groups, compared to the « Early Follicular Phase » defined as the reference group, we demonstrated the non-inferiority for LFP and OP groups, in the Per Protocole analysis and in the Intention To Treat analysis (Table 3 ). Table 3 Non Inferiority study, comparing the number of retrieved oocytes in the OP and EFP groups, in the per protocol (PP) and intention to treat (ITT) population Number of retrieved oocytes in the EFP group Mean (SD) Number of retrieved oocytes in the LFP (upper) and OP (lower) groups Mean (SD) Mean difference [95% two-tailed CI] One-sided p .value 5% PP population 13.1 (7.4) 15.8 (9.8) 18.7 (11.7) 5.9[1.6; 10.4] 5.6 [−0.2; 11.5] 0.002 0.018 ITT population 12.5 (7.7) 15.7 (9.8) 17.8 (11.4) 3.2 [−1.4; 7.9] 5.3[−0.01;10.6] 0.033 0.013 Non-inferiority is demonstrated by a one-sided P value<0.025 and a lower margin of the 95% confidence interval not surpassing the value −2. Non Inferiority study, comparing the number of retrieved oocytes in the OP and EFP groups, in the per protocol (PP) and intention to treat (ITT) population 15.8 (9.8) 18.7 (11.7) 5.9[1.6; 10.4] 5.6 [−0.2; 11.5] 0.002 0.018 15.7 (9.8) 17.8 (11.4) 3.2 [−1.4; 7.9] 5.3[−0.01;10.6] 0.033 0.013 Non-inferiority is demonstrated by a one-sided P value<0.025 and a lower margin of the 95% confidence interval not surpassing the value −2. Four minimal or mild OHSS cases were observed (4%), 1 in the EFP group, 1 in the MFP, 1 in the LFP, and 1 in the OP group in intention-to-treat analysis (Supplementary table II) with none requiring hospitalization. One premature LH peak was reported in the MFP group, with a peak at 10 IU/ml, and the donor received a GnRH antagonist injection the same day. An additional dose of follitropin beta was necessary for 95.5% of patients, significantly fewer in the LFG group (77.8%) than in the other groups ( p  = 0.005) in per-protocol analysis (Table II). The mean number of administered units was 702.1  ±  511.9. The median duration of stimulation was 10 days, with no significant difference between groups. The total median number of injections, including corifollitropin alpha and triggering, was five. Donor satisfaction was high (with a mean score of 9 on a scale of 0 to 10) with no significant difference between groups in per protocol analysis (Table II) and in intention-to-treat (Supplementary table II). Serum levels of estradiol, LH, and progesterone on the day of stimulation initiation (S1) were significantly different between the five groups. At S8, after the action of corifollitropin alfa, hormonal values remained significantly different, with higher level of estradiol and LH in the LFP group and higher progesterone levels in the LFP and OP groups (Fig. 3 ). However, the number of follicles measured at S8 did not differ between the five groups. Fig. 3 Serum levels of estradiol (3a), luteinizing hormone (LH), (3b) and progesterone (3c) measured on stimulation day 1 (S1) and day 8 (S8). a Comparative analyses of estradiol levels between periods S1 and S8 across groups, within the per-protocol population. b Comparative analyses of LH levels (UI/ml) between periods S1 and S8 across groups, within the per-protocol population. c Comparative analyses of progesteron levels (ng/ml) between periods S1 and S8 across groups, within the per-protocol population Serum levels of estradiol (3a), luteinizing hormone (LH), (3b) and progesterone (3c) measured on stimulation day 1 (S1) and day 8 (S8). a Comparative analyses of estradiol levels between periods S1 and S8 across groups, within the per-protocol population. b Comparative analyses of LH levels (UI/ml) between periods S1 and S8 across groups, within the per-protocol population. c Comparative analyses of progesteron levels (ng/ml) between periods S1 and S8 across groups, within the per-protocol population

This multicenter open-labelled pilot RCT was performed in four oocytes donation centers in France: Reproductive Medicine Department of Tenon University Hospital (Paris), Jean Verdier Hospital (Bondy), Clinique La Sagesse (Rennes) and Lille University Hospital (Lille) from July 2020 to September 2023. The study was approved by the Ethical Committee of Ile de France 2 (number of approval: 18.08.03.49811) and was prospectively registered at Clinical Trials. gov ( NCT 03895099 ). Eligible donors were aged 18 to 37 years, in accordance with the French Bioethics Law and national guidelines, which govern egg donation as a voluntary, altruistic and free act. Oocyte donors fulfilling the inclusion criteria and willing to participate in the study were recruited after signing informed consent for participation in a random-start PPOS protocol. The inclusion criteria were good general health, a normal karyotype, a family medical history free of inheritable disease, age between 18 and 37 years, an antral follicle count (AFC) > 8, a BMI between 18 and 32 kg/m2, a natural regular menstrual cycle, and no use of hormonal contraception for at least one cycle prior to oocyte donation. Participants also had to present no contraindications to oocyte donation. Candidates with severe endometriosis (stage 3–4 AFS), ovarian cyst of more than 3 centimeters, polycystic ovary syndrome, or contraindication of hormonal treatment were excluded. After presenting the research protocol and obtaining signed informed consent, participants were randomized into five groups according to the onset phase of ovarian stimulation initiation: Early Follicular Phase (EFP/D1-D3), Mid-follicular phase (MFP/D4-D7), Late follicular Phase (LFP/D8-D11), Ovulatory Phase (OP/D12-D15) or Luteal Phase (LP/D16-D30). Randomization was conducted electronically using randomization blocks, with stratification based on center and age (< 32 years or ≥ 32 years). Each participant received their prescriptions at the time of randomization, one to two menstrual cycles prior to the start of ovarian stimulation for donation in order to schedule endometrial preparation for recipients in the most frequent situation of synchronized cycles. A total of 110 oocyte donors were randomized (Fig. 1 ). An identical ovarian stimulation protocol was used for all groups (Fig. 2 ), with Desogestrel (75 mg, One tablet daily) administered from stimulation Day 1 to the day of triggering, and 150 µg Corifolitropin alfa (CFA) (Elonva 150, Organon) administered on stimulation Day 1, followed by 150 to 300 IU of recombinant follitropin beta (Puregon, Organon) from stimulation Day 8 if required, up to achieving triggering criteria, defined as the presence of at least three follicles measuring at least 17 mm in diameter. Fig. 1 Trial flowchart. In Intention to treat (ITT), participants were randomized into five groups according to the onset phase of ovarian stimulation initiation: Early Follicular Phase (EFP/D1-D3), Mid-follicular phase (MFP/D4-D7), Late follicular Phase (LFP/D8-D11), Ovulatory Phase (OP/D12-D15) or Luteal Phase (LP/D16-D30). In the per-protocol analysis (PP), 13 of the 102 patients were re-assigned to different stimulation onset groups based on their hormonal levels and ultrasound examinations at S1 (reviewed by investigators) Trial flowchart. In Intention to treat (ITT), participants were randomized into five groups according to the onset phase of ovarian stimulation initiation: Early Follicular Phase (EFP/D1-D3), Mid-follicular phase (MFP/D4-D7), Late follicular Phase (LFP/D8-D11), Ovulatory Phase (OP/D12-D15) or Luteal Phase (LP/D16-D30). In the per-protocol analysis (PP), 13 of the 102 patients were re-assigned to different stimulation onset groups based on their hormonal levels and ultrasound examinations at S1 (reviewed by investigators) Fig. 2 Protocol scheme. An identical ovarian stimulation protocol was used for all groups, with Desogestrel (75 mg, One tablet daily) administered from stimulation Day 1 to the day of triggering, and 150 µg Corifolitropin alfa (CFA) (Elonva 150, Organon) administered on stimulation Day 1, followed by 150 to 300 IU of recombinant follitropin beta (Puregon, Organon) from stimulation Day 8 if required, up to achieving triggering criteria Protocol scheme. An identical ovarian stimulation protocol was used for all groups, with Desogestrel (75 mg, One tablet daily) administered from stimulation Day 1 to the day of triggering, and 150 µg Corifolitropin alfa (CFA) (Elonva 150, Organon) administered on stimulation Day 1, followed by 150 to 300 IU of recombinant follitropin beta (Puregon, Organon) from stimulation Day 8 if required, up to achieving triggering criteria Final oocyte maturation was induced with a GnRH agonist (triptorelin 0.2 mg) (Decapeptyl Ipsen Pharma), followed by oocyte retrieval 34–36 h thereafter. We have performed blood samples (estradiol, LH, progesterone) and transvaginal ultrasound at stimulation Day 1, before Corifollitropin Alfa administration and Desogestrel onset (stimulation Day 1), at stimulation Day 8 and repeated if required. In case of LH surge (> 10 IU/ml), the decision was left to the clinician either to administer a GnRH antagonist (Ganirelix 0.25 mg, Orgalutran, Organon), injected daily up to the triggering day, or to cancel the cycle. The primary outcome was the number of oocytes retrieved after ovarian stimulation. The secondary outcomes included the number of mature oocytes retrieved, the occurrence of ovarian hyperstimulation syndrome (OHSS), classified according to the ASRM criteria and premature LH surge. We also compared hormone measurements (estradiol, LH, progesterone) and follicle count at stimulation day 8 (S8), number of injections and total dose of follitropin beta or other injections involved in the treatment, stimulation duration, and overall patient satisfaction with the treatment process. Satisfaction was assessed by phone 12 to 16 days after oocyte retrieval during the routine well-being follow-up call, using a 0-to-10 rating scale. This study was designed as a pilot study, and although the sample size was not formally calculated, it provides valuable preliminary data on the feasibility of the approach. Of the 110 randomized patients (Fig. 1 ), three refused the intervention, one did not start treatment, two were mistakenly included despite meeting exclusion criteria, one received incorrect ovarian stimulation treatment, and one was excluded due to missing informed consent. As a result, the modified intention-to-treat population included 102 patients who completed stimulation and underwent oocyte retrieval. Given this sample size of 102 participants across five groups, a post-hoc power analysis estimated that the study has over 60% power to detect a small effect size (Cohen’s f = 0.2) at a significance level of 0.05. While this level of power is lower than the conventional 80% threshold for confirmatory studies, it remains sufficient to generate meaningful insights. The observed effect size and variability estimates will inform the design of a subsequent adequately powered multicenter trial. Each individual was analyzed according to their assigned treatment arm. The planned date of corifollitropin alfa administration (S1) was determined at randomization based on the expected onset of the next menses in women with regular cycles. All participants received the injection on the scheduled date; however, physiological variations in cycle length led some to begin stimulation in a different menstrual phase than initially assigned. Consequently, in the per-protocol analysis, 13 of the 102 patients (12.7%) were reassigned by the investigators to different stimulation onset groups based on the presence of a dominant follicle and their hormonal profiles at S1. Patients with LH < 10 IU/L and progesterone < 3 ng/mL were classified as being in the early D1-D3 or mid-follicular phase D3-D7; those with progesterone  14 IU/L and progesterone < 3 ng/mL were considered to be in the ovulatory phase; and those with progesterone ≥ 3 ng/mL were classified as being in the luteal phase. Categorical variables were presented as counts and percentages, and continuous variables as mean (standard deviation) and median (interquartile range). Given the relatively small sample size and the non-normal distribution of most variables, group comparisons were primarily performed using the Kruskal–Wallis test, with a significance threshold of p  < 0.05; when normality assumptions were met, mean comparisons were conducted using ANOVA. For categorical variables, Pearson’s Chi-square or Fisher’s exact test was applied as appropriate. A post hoc power analysis indicated that the study had over 60% power to detect a small effect size (Cohen’s f = 0.2) at a 5% significance level. Accordingly, a non-inferiority analysis was conducted to assess whether the number of oocytes retrieved in the ovulatory phase was non-inferior to that of the reference group (EFP), using a non-inferiority margin of two oocytes, a one-sided alpha risk of 2.5%, and a 95% confidence interval of the mean difference in both the per-protocol (PP) and intention-to-treat (ITT) populations. Non-inferiority was concluded when p  < 0.025 and the lower limit of the 95% CI was greater than − 2. All analyses were performed using R software version 4.2.3.

Our pilot RCT study suggests that the total number of retrieved oocytes and the number of mature oocytes remain non statistically different regardless of the initiation day within the menstrual cycle when utilizing a ultra-flexible ovarian stimulation protocol designed for oocyte donors. This approach combines a PPOS protocol, with a random-start strategy with the use of corifollitropin alfa and GnRH agonist triggering. However, larger-scale studies are required to validate these findings before this protocol can be widely implemented.

This study evaluates the feasibility and efficacy of an ultra-flexible ovarian stimulation protocol tailored for oocyte donors. Specifically, we combined corifollitropin alfa with a progestin-primed ovarian stimulation (PPOS) approach, integrating a random-start protocol with GnRH agonist triggering. This is the first randomized, controlled, multicenter pilot study assessing the effectiveness of a PPOS protocol, stratified by the phase of ovarian stimulation initiation. Our pilot study demonstrates that the total and mature oocyte yields did not differ significantly according to the phase of stimulation initiation. Non-inferiority analysis further supported these findings for LFP and OP groups, addressing the limited power of conventional statistical testing due to the small sample size in each group. This finding aligns with results from conventional early follicular phase (EFP) initiation in oocyte donors. Numerous studies, particularly in the context of oncofertility, have shown that emergency ovarian stimulation can be successfully initiated at any phase of the menstrual cycle (random-start), yielding comparable outcomes in terms of oocyte retrieval between the luteal and follicular phases [ 9 , 10 ]. Similarly, Guerrero et al. [ 6 ]. found identical oocyte yields when ovarian stimulation was initiated in the early follicular, mid-late follicular, or luteal phases in oocyte donors. However, unlike our randomized study, Guerrero’s analysis was non-randomized. Recently in a randomized study, De Rijdt et al. [ 11 ]. also found a similar efficacy of early (D2) or late follicular phase stimulation (D13, before ovulation) in terms of number of oocytes retrieved. Beyond the random-start approach, we adopted the PPOS protocol to reduce the number of injections required and enhance flexibility in managing oocyte donors. Similar to the random-start strategy, PPOS protocols have gained popularity since 2015 [ 12 ] for ovarian stimulation without plans for fresh embryo transfers. Various progestins and administration routes have been explored, and Ata et al.’s [ 13 ] meta-analysis showed no significant differences in oocyte yield between PPOS and control groups. Notably, Martínez et al. [ 14 ] demonstrated comparable efficacy of PPOS versus the standard antagonist protocol in oocyte donors regarding oocyte retrieval. In a retrospective “before-and-after” study, Filippi et al. [ 15 ] compared a conventional antagonist protocol initiated in the early follicular phase with a PPOS random-start protocol in cancer fertility preservation cases. Their findings corroborated the equivalence of these protocols concerning cryopreserved oocyte yields. Regarding the efficacy of the PPOS protocol in suppressing LH surges, La Marca et al. [ 7 ] compiled data from 10 studies comparing PPOS to antagonist protocols and found no significant differences (0.7% vs. 0% LH surge occurrence). This aligns with our findings of a 1% LH surge rate. Furthermore, we did not observed any premature ovulation. A recent open-label randomized controlled trial comparing PPOS stimulation to GnRH antagonist protocol found similar results: no premature LH surge was observed in the PPOS group, while 3.5% of cases in the GnRH-ant group experienced premature LH surges, however, no premature ovulation occurred in either group. Furthermore, in this adequately powered study, the LBR for the first frozen embryo transfer in the PPOS group and the fresh transfer in GnRH-antagonist group were comparable [ 16 ]. In terms of safety, the combined PPOS–random-start protocol appears comparable to antagonist protocols, with lower rates of ovarian hyperstimulation syndrome (OHSS). For instance, Cui et al. [ 17 ] reported reduced OHSS rates with PPOS. Our study observed no severe OHSS cases and only a 4% incidence of mild OHSS. Our findings also reveal comparable gonadotropin consumption and stimulation duration across the five randomization groups. Although the luteal phase (LP) group exhibited slightly longer stimulation durations and higher gonadotropin consumption, these differences did not reach statistical significance. These results are consistent with Alexandru et al. [ 18 ] meta-analysis but contrast with studies by Boots et al. [ 19 ], Alexander et al. [ 20 ]. and Sönmezer et al. [ 21 ], which reported prolonged stimulation durations and increased gonadotropin requirements with luteal-phase initiation. As expected, serum estradiol, LH, and progesterone levels at stimulation initiation (S1) differed significantly across the groups. These hormonal differences persisted at S8, at the end of corifollitropin alfa action, with notably higher estradiol and LH levels in the LFP group and elevated progesterone levels in the LFP and OP groups. Despite these hormonal variations, follicle counts at S8 and oocyte yields remained consistent across groups. Interestingly, late follicular and ovulatory phase initiations were associated with significantly higher estradiol and LH levels at S8, with a nonsignificant trend toward higher oocyte retrievals. This raises the hypothesis that elevated LH and FSH levels from the ovulatory surge of gonadotrophins that occurred during the initial days of corifollitropin alfa action might enhance follicular recruitment and estradiol synthesis, potentially leading to increased oocyte yields. Larger studies are needed to confirm or refute the significance of this trend. Our secondary objective was to evaluate the feasibility of a donor-friendly protocol. Across all study groups, the number of injections was five in mean, and could be further reduced by utilizing nasal GnRH agonists for triggering rather than subcutaneous administration. Donors reported high levels of satisfaction (average score: 9/10) across all randomization groups. In alignment with the goal of implementing a patient-friendly protocol, we aimed to minimize the number of cycle monitoring visits to three - scheduled at stimulation day 1 (S1), stimulation day 8 (S8), and a final visit determined at the clinician’s discretion. This study is a proof-of-concept trial with a limited sample size but randomized, controlled, multicenter design. Larger population studies are necessary to confirm the efficiency of random-start protocols for oocyte donors. Additionally, while this trial focused on the stimulation initiation phases regarding total oocyte and mature oocyte yields, oocyte quality was not assessed. Recent studies, including Martinez et al. [ 22 ], have shown that ovarian stimulation initiated during the luteal or follicular phase yields a similar number of euploid embryos. Several studies [ 13 , 23 ] confirmed the equivalence of PPOS and antagonist protocols in fertilization, blastocyst formation, and implantation rates. Devesa et al. [ 24 ] also reported no differences in oocyte competence, embryo development, or live birth rates between PPOS and GnRH antagonist protocols using vitrified donated oocytes. Moreover, data comparing PPOS and GnRH antagonist protocols show no significant differences in embryo ploidy. La Marca et al. [ 25 ] found comparable rates of euploid embryos per injected oocyte between the two protocols, a finding confirmed by Wang et al. [ 26 ] in a larger cohort. An initial retrospective study by Chen et al. [ 27 ] found a significantly lower cumulative live birth rate (CLBR) and a longer time to pregnancy leading to live birth with PPOS compared to the GnRH antagonist protocol. However, regarding live birth rates and neonatal outcomes following frozen embryo transfers (FETs) from oocytes retrieved using PPOS versus antagonist protocols, most studies report no significant differences [ 16 , 26 , 28 ]. Additionally, congenital anomaly rates were found to be similar between the PPOS protocol (using chlormadinone acetate) and the antagonist protocol in a recent retrospective study by Shibasaki et al. [ 29 ].

Ovarian stimulation protocols have traditionally started with the administration of gonadotrophins in the early follicular phase, at the time of the inter-cycle FSH rise, leading to a growth of multiple follicles. This is associated with the blockage of premature LH surges by GnRH agonist or antagonist to avoid a premature ovulation and obtain a receptive endometrium for fresh embryo transfer. This conventional approach, which mimics physiology, is also widely believed to represent the optimal timing for multiple follicular recruitment. The documentation of multiple follicular cohorts or waves during the menstrual cycle challenged the traditional theory that a single cohort of antral follicles grows only during the follicular phase of the menstrual cycle [ 1 ]. It also provides the knowledge and physiological basis for the so-called ‘non-conventional’ ovarian stimulation approaches [ 2 ], such as the ‘random-start’ protocol, i.e., the initiation of the stimulation process irrespective of the phase of the natural menstrual cycle. The bulk of published studies on random-start ovarian stimulation involves women referred to oncologic units for oocyte cryopreservation [ 3 ] More recently, some studies have focused on the effectiveness of this strategy in patients undergoing elective cryopreservation of oocytes, including ‘social freezing’ [ 4 ] or infertile patients with deferred fresh embryo transfer (“freeze all” practice), both conditions that do not require a receptive endometrium [ 5 ]. Data from these publications suggest no difference in the number of oocytes obtained regardless of the day of the cycle on which stimulation is initiated compared to conventional protocols. To date, limited evidence is available on the use of this strategy in the context of egg donation, another target of patients who could benefit from random-start. Data from a recent retrospective study has demonstrated that the number of oocytes retrieved in donation cycles was comparable after conventional and random-start ovarian stimulation, and between subgroups, regarding the onset of the ovarian stimulation (>D4, D4-14 or >D14 up to D32 of the menstrual cycle) [ 6 ]. However the onset of ovarian stimulation was not assessed according to the phase of the menstrual cycle defined according to follicular growth and ovulation. Furthermore, recent studies have also demonstrated the efficacy and safety of the Progestin-primed ovarian stimulation (PPOS) protocol in oocyte donors. A large review [ 7 ] has recently demonstrated the ability of progesterone and its derivatives to block the LH surge. PPOS is shown to effectively inhibit spontaneous ovulation, without affecting the number of retrieved oocytes and embryo quality. The use of progestins allows better control of LH concentrations, lower costs and easier administration by oral route compared to injections required for GnRH antagonist administration. Recent data have confirmed the efficacy of Desogestrel to block the LH surge in a sample of oocyte donors, providing a more “friendly” protocol [ 8 ]. In order to provide a more user-friendly, simplified and flexible ovarian stimulation protocol that reduces the burden for donors, we studied the impact on oocyte yield of combining a random-start corifollitropin alfa stimulation with the concomitant prevention of premature LH surge by progestin administration in a PPOS protocol with GnRH agonist triggering. The aim of this feasibility pilot randomized control study was to assess the equivalence in term of number of retrieved oocytes of a delayed onset of corifollitropin alfa stimulation with concomitant administration of desogestrel in Mid Follicular Phase (MFP/D4-D7 of the natural cycle), Late Follicular Phase (LFP/D8-D11), Ovulatory Phase (OP/D12-D15) or Luteal Phase (LP/D16-D30) compared to the standard start in Early Follicular Phase (EFP/D1-D3).

Supplementary Material 1 Supplementary Material 1 Supplementary Material 2 Supplementary Material 2