Subcutaneous progesterone administration provides a similar ongoing pregnancy rate compared with intramuscular progesterone administration in hormone replacement therapy frozen embryo transfer cycles.

During the study period, 444 cycles of FET with HRT were performed. Of these, 224 patients were eligible for the study, and 133 were included in the SC-P group and 91 in the IM-P group ( Fig. 1 ). Table 1 presents the demographic, cycle, and embryologic characteristics of the groups. The baseline characteristics were similar between groups. The percentages for quality and vitrification day (day 5 to day 6) of transferred blastocysts were comparable between the 2 groups. The median circulating P levels on the day of ET was 19.92 (15.19-27.23) ng/ml and 21 (16.5-27.23) ng/ml in the SC-P and IM-P groups, respectively ( P =.786). The clinical pregnancy rates (CPRs) (86/133 [64.7%] vs. 57/91 [62.6%]; P = .757), miscarriage rates (MRs) (21/86 [24.4%] vs. 10/57 [17.5%]; P =.328), and OPRs (65/133 [48.9%] vs. 47/91 [51.6%]; P =.683) were comparable between the SC-P and IM-P groups ( Table 2 ). Binary logistic regression was performed for OP as the dependent factor, and woman’s age, BMI, previous live births, previous attempts, previous miscarriages, day of vitrification (day 5 or 6), ET-day P levels, embryo quality, and P administration route (SC-P and IM-P) were chosen as independent factors. The embryo quality was a significant independent prognosticator (for poor quality embryos aOR: 0.11, 95% CI, 0.029-0.427, P =.001) ( Table 3 ). Binary logistic regression revealed that the P route (SC-P vs. IM-P) was not a significant factor for OP ( P =.286). Table 1 Comparison of patient, cycle and embryologic characteristics of the SC-P and IM-P-treated groups. Variables SC-P (n = 133) IM-P (n = 91) P value Female age (y) 29 (28-32) 30 (28-31) .855 BMI (kg/m 2 ) 23.66 (21.02-27.92) 22.15 (20.55-25) .058 No. of previous live births (median Q1-Q3) 0 (0-0) 0 (0-0) .892 No. of previous miscarriages (median Q1-Q3) 0 (0-0) 0 (0-0) .436 No. of previous attempts (median Q1-Q3) 0 (0-1) 0 (0-1) .023 Duration of infertility (median Q1-Q3) (years) 5 (3-10) 5 (4-9) .091 AFC (median Q1-Q3) 16 (11-26) 15 (11-25) .391 Main diagnosis .978  Male factor 18/133 (13.5) 12/91 (13.2)  Polycystic ovary syndrome 26/133 (19.6) 18/91 (19.8)  Endometriosis 6/133 (4.5) 4/91 (4.4)  Tubal factor 8/133 (6) 5/91 (5.5)  Unexplained 35/133 (26.3) 26/91 (28.5)  Diminished ovarian reserve 6/133 (4.5) 3/91 (3.3)  Combined 34/133 (25.6) 23/91 (25.3) E 2 levels on P administration day (pg/ml) 232 (180-334) 234 (196-318) 1 P levels on P administration day (ng/ml) 0.20 (0.13-0.31) 0.16 (0.07-0.24) .051 Endometrial thickness (mm) 10 (8.70-11) 10 (9-11) .293 P levels on the day of ET (ng/ml) 19.92 (15.19-27.23) 21 (16.5-27.23) .786 Embryo quality (%)  Good 81/133 (60.9) 55/91 (60.4) .273  Fair 32/133 (24.1) 28/91 (30.8)  Poor 20/133 (15) 8/91 (8.8) Day of transfer (%)  Day 5 112/133 (84.2) 71/91 (78) .239  Day 6 21/133 (15.8) 20/91 (22) Values are given as the median (Q1-Q3) or number (percentage)
 BMI = body mass index; AFC = antral follicle count; MII = mature oocyte, 2PN = 2 pronuclei, E 2 = estradiol, P = progesterone, ET = embryo transfer. P values were calculated with a χ 2 test and a nonparametric independent samples median test. Table 2 Comparison of pregnancy outcomes between SC-P and IM-P. SC IM P value Clinical Pregnancy Rate (n/%) 86/133 (64.7) 57/91 (62.6) .757 Ongoing Pregnancy Rate (n/%) 65/133 (48.9) 47/91 (51.6) .683 Miscarriage Rate (n/%) 21/86 (24.4) 10/57 (17.5) .328 SC = subcutaneous; IM = intramuscular; P = progesterone. Table 3 Binary logistic regression analysis for ongoing pregnancy in all study patients. Variables Unadjusted (Crude) OR 95% CI for Crude OR P value Adjusted OR 95% CI for Adjusted OR P value Age (y) 0.985 [0.908-1.068] .71 1.002 [0.907-1.108] .963 BMI (kg/m 2 ) 0.951 [0.896-1.011] .106 0.956 [0.894-1.022] .186 Previous live births 1.831 [0.824-4.071] .138 1.717 [0.627-4.702] .293 Previous attempts 0.866 [0.681-1.101] .241 0.999 [0.698-1.429] .996 Previous miscarriages 0.872 [0.60-1.268] .473 0.753 [0.475-1.195] .229 Embryo quality  Good (Reference) .006  Fair 0.819 [0.445-1.507] .521 0.793 [0.378-1.662] .539  Poor 0.128 [0.042-0.388] <.001 0.111 [0.029-0.427] .001 Endometrial Thickness (mm) 1.079 [0.925-1.258] .334 1.158 [0.962-1.395] .121 ET-day P level (ng/ml) 1.019 [0.997-1.042] .097 1.011 [0.983-1.039] .442 Embryo day of vitrification  Day 5 (Reference)  Day 6 0.45 [0.222-0.913] .027 0.661 [0.274-1.594] .357 P administration route  SC-P (Reference)  IM-P 1.117 [0.655-1.905] .683 0.694 [0.354-1.358] .286 Constant 0.823 - .923 BMI = body mass index; IM-P = intramuscular progesterone; SC-P = subcutaneous-progesterone. Comparison of patient, cycle and embryologic characteristics of the SC-P and IM-P-treated groups. Values are given as the median (Q1-Q3) or number (percentage) BMI = body mass index; AFC = antral follicle count; MII = mature oocyte, 2PN = 2 pronuclei, E 2 = estradiol, P = progesterone, ET = embryo transfer. P values were calculated with a χ 2 test and a nonparametric independent samples median test. Comparison of pregnancy outcomes between SC-P and IM-P. SC = subcutaneous; IM = intramuscular; P = progesterone. Binary logistic regression analysis for ongoing pregnancy in all study patients. BMI = body mass index; IM-P = intramuscular progesterone; SC-P = subcutaneous-progesterone. In the SC-P and IM-P groups, ET-day P levels were divided into quartiles (Q) to evaluate the effect on pregnancy outcome ( Supplemental Table 1 , available online). The serum P intervals for each quartile for SC-P were: Q 1 27.23 ng/ml. The OPRs were 33.3% (11/33), 50% (17/34); 60.6% (20/33); and 51.5% (17/33) in Q 1 , Q 2, Q 3 , and Q4, respectively. ( P =.163). The MRs were 26.7% (4/15), 26.1% (6/23), 20% (5/25) and 26.1% (6/23) for Q 1 , Q 2 , Q 3 , and Q 4 , respectively ( P =.945). The serum P intervals of the IM-P group were 27.23 ng/ml in Q 1 , Q 2, Q 3 , and Q4, respectively. In contrast, in the SC-P group, the OPR of Q 1 was significantly reduced compared with that of Q 2 , Q 3 , and Q 4 (26.1% [6/23], 65.2% 15/23], 54.5% [12/22] and 60.9% [14/23], respectively, P =.031). The MR was significantly higher in Q 1 than in Q 2 , Q 3 , and Q 4 (50% [6/12], 16.7% [3/18], 0% [0/12], and 6.7% [1/15], respectively, P =.03) (SupplementalTable 1). In the SC-P group, when OP was chosen as the dependent factor, and when woman’s age, BMI, previous live births, previous attempts, previous miscarriages, embryo quality, day of vitrification (day 5 or 6), endometrial thickness, and ET-day P levels were chosen as independent factors, embryo quality was found to be a significant independent factor for OP (aOR for poor quality, 0.099; 95% CI 0.019–0.507, P =.006) ( Supplemental Table 2 ) Logistic regression analysis revealed that in addition to embryo quality ( P =.048), the ET-day P level was an independent factor for OP in the IM-P group (Q 1 vs. Q 2 +Q 3 +Q 4 ; aOR, 8.178; 95% CI, 1.387–48.223, P =.02) ( Supplemental Table 3 ).

The protocol of this nonrandomized prospective study was approved by the Üsküdar University Ethics Committee (Number: 61351342; date: June 30, 2020). The study was conducted at Bahceci Fulya IVF Center between July 1 and October 31, 2020. A total of 444 patients who started endometrial preparation for FET were assessed for eligibility during this period ( Fig. 1 ). Because we aimed to compare homogeneous patient groups, only the first FET cycle was included after freeze-all for each patient. The indications for the freeze-all approach were ovarian hyperstimulation syndrome and a high P level (>1.5 ng/ml) on the trigger day. The exclusion criteria were as follows: women aged >35 years, PGT-A cycles, cleavage-stage ET, more than one ET, patients with uterine pathology, patients with hydrosalpinx, FET with surplus embryos, P supplementation with the vaginal route or vaginal route in combination with IM route, natural cycle FET and endometrial thickness <7 mm on the day of P administration. Figure 1 Flowchart of the study Flowchart of the study The oocyte retrieval, denudation, and intracytoplasmic sperm injection (ICSI) procedures were performed as previously described ( 21 ). The ICSI was used in all cases per the established policy in our country. After microinjection, oocytes were cultured individually in a special, pre-equilibrated culture dish. All embryos were kept in benchtop incubators (MIRI, ESCO Medical, Singapore) and cultured until day 6 of embryo development. Details on embryo development were recorded daily. Blastocyst grading was performed at 114 hours (day 5) and 138 hours (day 6) after ICSI, as described by Gardner and Schoolcraft ( 22 ). Embryos that reached the blastocyst stage were frozen by vitrification. All embryos in this study were vitrified and warmed with commercial vitrification and warming kits (Vit Kit 1 -Freeze, 90133-SO, Irvine Scientific; Vit Kit 1 -Thaw, 90137-SO, Irvine Scientific) using an open carrier device (Cryotech, Reprolife, Japan). The freezing and warming processes were performed according to the manufacturer’s instructions, as previously described. After warming, all embryos were transferred to and maintained in an equilibrated culture dish before uterine replacement. To perform proper grading, a blastocyst evaluation was performed 2–3 hours after warming was completed. The endometrial preparation for ET involved HRT. Oral estrogen (Estrofem, Novo Nordisk, Turkey) administration was used according to a step-up regimen: 4 mg/day on days 1–4, 6 mg/day on days 5–8, and 8 mg/day on days 9–12. Endometrial thickness was measured by transvaginal ultrasonography (TV-USG) on days 10–13, and if the endometrial thickness was 1.5 ng/ml, the cycle was canceled. According to the patient’s accessibility to the hospital and preference, P was initiated either with 50 mg IM-P injection (Progestan, Kocak Farma, Turkey) once daily or with 25 mg SC-P (Prolutex, IBSA, Switzerland) injections twice daily. The first dose of IM-P was injected at 4 PM, and subsequent doses were repeated every 24 hours at the same time interval. For the SC-P injection, the first dose was injected at 4 PM and continued twice daily at 8:00 AM/8:00 PM. All transfers were performed between 3 PM and 6 PM under ultrasonographic guidance. Serum β-human chorionic gonadotropin (hCG) levels were measured 12 days after FET, and levels ≥5 IU were accepted as positive. Oral estrogen replacement was stopped at the 6th week of pregnancy, whereas P was continued until the10th week in both arms. Blood samples were obtained to determine serum P on the 6th day of P administration, at 3 PM, which was the 119th hour after P initiation. Embryo transfer day was preferred to maximize patient compliance for the study. Serum P concentrations were measured by an electrochemiluminescence immunoassay (Cobas Elecsys Progesterone III, Roche Diagnostics GmbH, Germany). The intra-assay coefficient of variation was 2.4%, and the interassay coefficient of variation was 3.9%. The sensitivity of the assay was 0.03 μg/l. Clinical pregnancy was defined as the detection of an intrauterine gestational sac by TV-USG, and OP was defined as a viable pregnancy detected by ultrasound examination at 12 weeks of gestation. Miscarriage was defined as loss of clinical pregnancy before gestational week 12. To calculate the sample size required for this study, R software program version 4.1.1 was used with SampleSize4ClinicalTrials package version 0.2.3. The aim of the study was to look at the difference in the proportion of ongoing pregnancy rates between the 2 treatment groups, namely, IM-P and SC-P. Therefore, the design of the study used was a parallel, noninferiority trial. The alpha level used was 5%, the power of the study used was 90%, the positive response (having OP) rate used was 50% in both IM-P and SC-P groups, and the noninferiority limit used was 20%. The results of this power analysis suggest that if there is truly no difference between IM and SC groups (50% in both groups), then 216 patients are required to be 90% sure that the upper limit of a one-sided 95% confidence interval will exclude a difference in favor of the IM group of < 20%. Also, if approximately 3.5% loss to follow-up is taken into account, extra 8 patients can be taken, which increases the total sample size to 224 patients. Thus, 224 patients were taken into consideration in this nonrandomized prospective study. The patient decided the route of P administration according to their preference and accessibility to the hospital. In the SC-P group consisting of 133 patients, the subcutaneous route for P was used in the HRT-FET cycles, whereas in the IM-P group consisting of 91 patients, the intramuscular route for P was used in the HRT-FET cycles. Kolmogorov-Smirnov and Shapiro-Wilk tests were used to determine the continuous quantitative parameters measured in the study. It was concluded that none of the continuous quantitative parameters investigated in the study followed a normal distribution. All tables, median and corresponding Q 1 (25th percentile) and Q 3 (75th percentile) values are reported for each specific quantitative measurement. The independent median test was used to compare those characteristics measured as quantitative measurements between the IM-P- and SC-P-treated groups, and Pearson’s χ2 test was used to demonstrate the differences in the proportion of those characteristics given as qualitative variables. The pregnancy rate, ongoing pregnancy rate (OPR), and miscarriage rate per patient in the SC-P and IM-P groups were further compared according to quartiles of serum P concentrations on the ET day using Pearson’s χ2 test. The range of the serum P levels on the day of ET was narrowed by examining quartiles, namely, the 25th, 50th, and 75th percentiles, respectively Q 1 , Q 2 , Q 3, and Q 4 (0–25% was Q 1 , 26–50% was Q 2 , 51–75% was Q 3 , and 76–100% was Q 4 .) Statistical analysis was conducted using the SPSS 25.0 software package (IBM, 2016). A P value P ≤.05 was considered to indicate statistical significance for all statistical tests. Binary logistic regression models were computed to model the outcome of having an OP in the entire study group, the SC-treated, and the IM-treated group. Three models were performed separately for all patients, the SC-treated, and the IM-treated group, to establish the effects of numerous factors in women on the likelihood of having an OP: age, body mass index (BMI), number of previous live births, previous miscarriages, attempts to conceive, embryo quality, endometrial thickness, day of ET and whether the patient was in the ET-P-Q 1 group. In both models, the entering procedure was used to calculate the adjusted odds ratios (aORs) and corresponding P values for each variable.

This nonrandomized prospective study demonstrated that patients with SC-P administration had similar pregnancy outcomes to those with IM-P in HRT-FET cycles. In the SC-P group, ET-day serum P level does not affect OPR, whereas, in the IM-P group, the OPR was found to be significantly lower in the low ET-day serum P level. The efficacy of SC-P for luteal phase support after fresh ETs was evaluated in 2 randomized control trials, and there was no significant difference in the pregnancy outcomes ( 6 , 7 ). Both randomized control trials compared SC-P with vaginal P. There is little agreement on the best route of P administration for P replacement in the HRT-FET cycles ( 1 ). In a recent three-armed randomized control trial, only the vaginal P arm had a significantly reduced pregnancy outcome compared with daily IM-P and vaginal P plus IM administration every 3 days ( 4 ). However, IM-P administration causes side effects and severe discomfort. Moreover, in our country, IM-P administration is allowed to be done under professional medical observation. Only 2 retrospective studies evaluate the efficacy of SC-P in the HRT-FET cycles. Turkgeldi et al compared SC-P with vaginal P in frozen-thawed blastocyst transfers and reported that pregnancy outcomes were similar (39.3% for SC-P versus 35.5% for vaginal P; P =.57)( 10 ). However, in this study, multivariate regression analysis was not performed. A previous retrospective analysis by our group showed that the live birth rate of SC-P was similar to that of IM-P for blastocyst frozen-thawed blastocyst transfers ( 9 ). Live birth rates were equivalent between the 2 treatments (62.7% vs. 58.7% in the SC-P and IM groups, respectively, P =.404). The P route was a nonsignificant factor after binary logistic regression. To our knowledge, this is the first nonrandomized prospective study to evaluate the efficacy of SC-P compared with IM-P in the HRT-FET cycles, and the finding was that there was no significant difference between SC-P and IM-P administration in terms of pregnancy outcomes (CPR; 64.7% vs. 62.6%, P =.757, MR; 24.4% vs. 17.5%, P =.329 OPR; 48.9% vs. 51.6%, P =.683). The P route was a nonsignificant factor for OP after binary logistic regression. Our study shows that SC-P administration effectively supports the luteal phase in the HRT-FET cycles. Because there is no corpus luteum formation in an HRT-FET cycle, exogenous administration P is warranted. There is no agreement in the literature on the optimal dose of progesterone for luteal phase support. It is reported that both 25 mg and 50 mg daily doses of SC-P administration led to decidualization in reproductive-aged women who were down-regulated with GnRH agonists and treated with estradiol to develop the endometrium ( 23 ). The serum P levels observed in this study were low compared with observed midluteal levels in a spontaneous menstrual cycle in the 25 mg arm (5.77 ng/ml) ( 24 ). There is no data on the optimal dosage of P in the various forms of P routes. Although the clinical effectiveness of 25 mg SC-P is demonstrated in 2 RCTs for fresh embryo transfers where multiple corpus luteum exist, the clinical effectiveness of 25 mg SC-P in HRT-FET cycles is not reported in any clinical study. All retrospective studies comparing the effectiveness of SC-P to vaginal P and IM-P route used 50 mg SC-P (25 mg bid). For the reasons mentioned above, administering 50 mg SC-P seems more convenient in achieving sufficient luteal support. It could also be speculated that a smaller dose may provide fewer side effects at the injection site by splitting the 50 mg dosage twice daily and may improve patient compliance. However, there is no data on the effect of splitting dosage on the pharmacokinetic and pharmacodynamic parameters. The HRT-FET cycles are more frequently preferred because of the advantages of scheduling the day of ET and having less monitoring. After the administration of estrogen, starting from the second day of the menstrual cycle for 10–13 days, P supplementation is started, mimicking the mid-cycle shift to the secretory phase of the cycle and preparing the endometrium for implantation. At this point, there is no corpus luteum, and a luteal phase must be created. The monitoring of the luteal phase in HRT-FET cycles has been investigated over the last 7 years. In the natural cycle, P levels secreted from the corpus luteum fluctuate ( 25 ). Although this fluctuation makes it hard to assess a threshold in spontaneous conceptions, the estimated threshold was found to be approximately 9.4 ng/mL ( 24 ). Previous studies of HRT-FET demonstrated the correlation between serum P and pregnancy outcomes with vaginal P supplementation ( 11 , 12 , 13 , 14 , 15 , 17 , 26 ). The most recent large-scale prospective study using vaginal P administration reported that serum P levels <8.8 ng/ml on the day of ET lowered the OPR in both own and donated oocyte cycles for HRT endometrial preparation ( 15 ). The serum P concentrations achieved after vaginal P administration are lower than those achieved with systemic P administration. However, endometrial tissue P concentrations are higher after vaginal P administration ( 26 ). Although serum P levels do not reflect tissue concentrations, studies have reported that lower values are correlated with worse pregnancy outcomes. Pharmacodynamic and pharmacokinetic studies reported that after 6 doses of P by vaginal administration, steady serum P levels are reached and that the levels of vaginal P are more stable ( 27 ). Although serum P level fluctuations are observed after IM-P administration, our recent prospective study showed that there is a minimum threshold of serum P that needs to be reached when using IM-P to obtain better results. In parallel with our previous study ( 18 ), this study revealed that the OPR for ET-day P levels < 16.5 ng/ml was significantly lower. In addition, ET-day P levels were found to be an independent factor for OP in the logistic regression analysis. Aqueous P preparation has the advantage of being well tolerated after SC-P administration. In pharmacokinetic and pharmacodynamic studies of this new product in which the same dosage of P was administered, the areas under the curve for the SC-P aqueous formulation and IM-P in oil formulation were comparable ( 5 , 28 ). The maximum serum concentrations of SC-P were 34 times higher than those of IM-P. Moreover, the time to achieve the maximum concentration was 7 times shorter in the SC-P group. In our study, the median circulating P levels of the SC-P and IM-P groups on the day of ET were similar ( P =.786). There were no significant differences in OPR for the SC-P group among Q 1 , Q 2 , Q 3, and Q 4 . In addition, multivariate logistic regression showed that the ET-day P level was a nonsignificant parameter for OP in the SC-P group. The difference in pharmacokinetic and pharmacodynamic properties of IM-P and SC-P may lead to steady concentrations, which makes ET-day P levels nonsignificant. The shorter time to achieve the maximum concentration in SC-P administration may cause quicker steady concentration, and this pharmacokinetic difference may lead to ET-day P levels being nonsignificant in the SC-P administration. Comparative studies evaluating the factors that may affect the pharmacokinetic and pharmacodynamic properties in the IM-P and SC-P administration are warranted for future perspectives. From another perspective, it can be speculated that twice a day injection in SC-P may influence steady concentration. However, there is no pharmacokinetic study that may support this hypothesis. Monitoring the luteal phase in HRT-FET cycles is being investigated to develop an algorithm for personalized luteal phase support. In a recent study, body weight, age, time of blood sampling, and a history of low P were associated with P concentrations before blastocyst FET ( 29 ). This study also reported that P concentrations vary during the day, even when exogenous hormone replacement is given. Interpersonal variations in serum P levels were euated by using the same luteal phase support, and significant variability was reported ( 30 ). Future studies, evaluating the factors affecting serum P levels after IM-P and SC-P administration, are mandatory for the individualization of luteal phase support for the HRT-FET cycles. Although we have information on the safety and efficacy of subcutaneous progesterone, little is known about patients’ acceptance and opinions about its administration. In a recent study, patient experience was evaluated in comparison with the recent SC-P administration to previous vaginal P administration, and the study reported SC-P is comfortable, easy to use, and leads to a high level of satisfaction among patients who had previously used vaginal progesterone ( 31 ). In our study, we did not assess the degree of acceptance and satisfaction. However, patients that we observe in our clinical practice complain of less pain in SC-P compared with IM-P. The HRT-FET cycles have an increased risk for obstetric, perinatal, and postpartum outcomes compared with natural cycle FET ( 32 ). The absence of corpus luteum causes decreased serum relaxin and vascular endothelial growth factor levels, lower reactive hyperemia index, and lower angiogenic circulatory endothelial progenitor cells. These factors play a role in the suggested mechanism underlining increased risks for obstetric, perinatal, and postpartum outcomes. Our study did not evaluate live birth, perinatal or postpartum outcomes. However, our total miscarriage rate in the study group was 21% (21/143) which was higher than expected. Future studies evaluating the effects of progesterone type on miscarriage rate and perinatal outcomes are warranted. This is the first prospective study evaluating the efficacy of SC-P administration in the HRT-FET cycles compared with the IM-P route. In addition, the study population has homogenous characteristics, which eliminates possible bias and has made the study more reliable for comparing these 2 groups. The main limitation of our study is that it is not a randomized controlled study. Larger randomized controlled studies following this preliminary study are needed to evaluate the efficay of SX-P administration in HRT-FET cycles. Another limitation is that we are able to report the OPR but not the live birth rates. In addition, only one blood sample was evaluated in the study. The effect of serum P levels on the day of ET in HRT-FET cycles should be evaluated in women with a wide range of characteristics. The inter- and intrapatient variability assessment will highlight the main domains for individualizing the luteal phase support in the HRT-FET cycles. We did not perform a cost analysis for the products used in our study. The daily medication prices are higher in the SC-P group. However, a cost-effectivity analysis should include all variables related to the pregnancy period and be considered, similar to the outcomes and obstetrical problems. This study did not include obstetrical outcomes. Another limitation is the lack of information about side effects and indirect health-related expenditures like transportation and loss of work because of the intervention for both IM-P and SC-P groups. In conclusion, SC-P administration seems to be an effective alternative to the IM-P route in patients with HRT-FET. Studies including a wide range of patients (large-scale BMI and woman’s age) are warranted to better define the effects of ET-day P levels on SC-P administration in HR-FET.