Efficacy of four vaginal progesterones for luteal phase support in frozen‐thawed embryo transfer cycles: A randomized clinical trial

Abstract Purpose To investigate the efficacy of four vaginal progesterones, Lutinus, Utrogestan, Luteum, and Crinone, as luteal phase support (LPS) in frozen‐thawed embryo transfer (Frozen‐ET) cycles. Methods Patients undergoing autologous Frozen‐ET of one cleavage‐stage embryo or one blastocyst. Two hundred fifty‐nine Frozen‐ET cycles were randomized to four groups for LPS: Lutinus, Utrogestan, Luteum, and Crinone. The clinical pregnancy rate (CPR), fetal heartbeat rate (FHR), and miscarriage rate (MR) were analyzed using the Mann‐Whitney or Kruskal‐Wallis test and Fisher exact test. Results Two hundred thirty‐five Frozen‐ET cycles were analyzed: 63 cycles in the Lutinus group, 60 in the Utrogestan group, 56 in the Luteum group, and 56 in the Crinone group. No significant differences were observed between the four groups in CPR (Lutinus, Utrogestan, Luteum, and Crinone: 34.9%, 33.3%, 37.5%, and 35.7%, respectively; P = .976), FHR (26.9%, 31.6%, 30.3%, and 25.0%, respectively; P = .857), and MR (31.8%, 10.0%, 19.0%, and 30.0%, respectively; P = .306). Multivariate logistic regression analysis also revealed that there were no statistically significant differences between the four groups with regard to CPR, FHR, and MR. Conclusion There was no clinically significant difference in pregnancy outcomes between the four vaginal progesterone groups for LPS in Frozen‐ET cycles.

menstrual cycle for the purpose of another pregnancy. In addition, freezing embryos and storing them eliminates the need for the patient to undergo repeated egg retrieval procedures; therefore, it allows the endometrium to maintain the normal hormonal environment, which increases the pregnancy rate. 4,5 The use of the Frozen-ET technique also avoids the risk of ovarian hyperstimulation syndrome.
It is a well-known fact that luteal phase support (LPS) in ART is essential for the implantation and maintenance of pregnancy. 6 However, in contrast to Fresh-ET, the absence of endogenous serum progesterone (P4) secretion before frozen embryo transfer in hormone replacement therapy cycles (HRT-FET) results in the need for exogenous P4 formulations at approximately pregnancy weeks 8 to 10. [6][7][8] Thus, it is important to identify the dose of vaginal preparation alone that will facilitate continuation of the pregnancy.
There are two Frozen-ET methods. Frozen-ET that utilizes the patient's natural ovulation cycle does not involve the use of hormone administration, which reduces the burden placed on the patient. However, as the day of ovulation must be identified, this method requires that the patient be examined frequently, and it does not allow freedom in selecting the day on which the transfer is to take place. On the other hand, HRT-FET, ovulation, and corpus luteum formation typically do not occur, and endometrial preparation requires exogenous P4 replacement. Since this method requires that the endometrium be artificially prepared, it gives patients the freedom to choose the date the transfer is to take place and keeps patient hospital visits to a minimum. Although there have been meta-analysis studies done on both natural ovulation cycle with Frozen-ET and HRT-FET, they did not find significant differences in terms of the clinical pregnancy rate (CPR), ongoing pregnancy rate (OPR), or delivery rate. 9 As a result, HRT-FET seems to be preferable among working women.
Methods for administering P4 formulations in HRT-FET are divided into intramuscular, oral, and vaginal administrations. Oral preparations, however, are subject to liver metabolism and are avoided owing to poor bioavailability and inferior pregnancy outcomes. 10 Intramuscular injection is highly effective, but self-administration is difficult and intense pain occurs at the site of injection.
The long-term use of intramuscular injection poses a serious burden, so patient satisfaction is higher with vaginal preparations. 11 Many studies have shown the effect of vaginal supplements in Fresh-ET cycles, in which vaginal P4 preparations and intramuscular injections have equivalent pregnancy rates. [12][13][14] For these reasons, many medical facilities have shifted from intramuscular injections to vaginal preparations. In fact, according to a 2012 online survey, the usage rate of vaginal preparation as monotherapy is 77%, making it the most popular method. 15 Vaginal P4 has been used for a long time in many countries, and many reports have shown its effectiveness in Fresh-ET. 12,14,[16][17][18][19][20][21] Although vaginal preparations are effective when used in Fresh-ET according to a systematic review in 2018, 22 a review in 2014 found that the data for their use in HRT-FET were insufficient and the timing of LPS, method of application, and dose were all unknown. 6 These are serious issues given that the number of Frozen-ET procedures performed worldwide is increasing.
In Japan, unlike overseas, oral, in-hospital vaginal, or intramuscular injection P4 preparation has long been used for luteal support in ART. In 2014, the P4 vaginal agent Lutinus was first approved, and in 2016, Utrogestan, Luteum, and Crinone were released successively. We conducted a prospective, randomized controlled trial with HRT-FET in 2018 using three vaginal P4 suppositories: Lutinus, Utrogestan, and Luteum, and found no significant difference between the three groups in the CPR, OPR, and miscarriage rate (MR). 23 During that study, Crinone was released, so this time, we conducted a new prospective, randomized comparative study on the use of four vaginal P4 suppositories in HRT-FET, Lutinus, Utrogestan, Luteum, and Crinone, in order to investigate CPR, fetal heartbeat rate (FHR), and MR.

| Study population
As this study was an exploratory study, no particular sample size was established. We excluded patients who had contraindications that were described on the drug package. This was the only exclusion criterion; we used no other restrictions such as age or cause of infertility. Thus, we requested the participation of all patients who visited our clinic on an outpatient basis, underwent egg retrieval, and had at least one embryo frozen using the vitrification method for cryopreservation. Patients who consented to study participation were enrolled in the study. The patients underwent autologous Frozen-ET of one cleavage-stage embryo or one blastocyst. Patients who desired to transfer two embryos or those who desired the use of a different preparation method were not permitted to participate in this study. Patients were permitted to participate two or more times.
Even if the patients participated in the study more than once, we allocated them to any of the four drug groups. Therefore, sometimes patients were assigned to the same vaginal drug used previously.

| Randomization and intervention
All patients were given a registration number on the basis of the order of their referral. Then, a specific researcher generated a computer-based random allocation table, and all patients were randomly assigned to one of the four study groups.
Those assigned to the first study group received 100 mg of a vaginal P4 tablet three times daily (Lutinus, Ferring Pharmaceuticals); those assigned to the second group received 200 mg of vaginal P4 capsules (Utrogestan, FUJIFILM Pharmaceuticals) three times daily; those assigned to the third group received a 400-mg vaginal suppository (Luteum, ASKA Pharmaceutical) twice daily; and those in the fourth group received 90 mg of a vaginal gel (Crinone, Merckserono) once daily. The randomization lists were kept on a password-protected computer.

| Study protocol
All patients underwent an IVF cycle with an agonist, antagonist, and mild stimulant protocol. IVF, intracytoplasmic sperm injection (ICSI), or both methods were used in the process. Confirmation of fertilization was tested for at 16-19 hours after IVF or ICSI. Several embryologists affiliated with the clinic performed cryopreservation of early embryos 2-3 days after egg retrieval and cryopreservation of blastocysts 4-7 days after egg retrieval. The vitrification method of cryopreservation was used in all cases. Before freezing, the early embryos were assessed using the Veek method, 24 and the blastocysts were assessed using the Gardner method. 25 In the case of several early embryos and blastocysts, the decision regarding whether to freeze them was left to four physicians affiliated with this clinic.
The HRT protocol was conducted using the gonadotropin-releasing hormone agonist via nasal drops for down-regulation 1-2 weeks before the start of menstruation. We started with the use of estrogen (E2) tape (Estrana tape, Hisamitsu Pharmaceutical Co., Inc.) or E2 gel (L'estrogel 0.06%, FUJIFILM Pharmaceuticals), which is a transdermal E2 preparation, from menstrual cycle days 2-5, gradually increased the dosage, checked whether the endometrial thickness was at least 7.0 mm on menstrual cycle days 12-15, and then started P4 vaginal suppositories (day 0), with embryo transfer being performed on days 2 and 3 for a cleavage-stage embryo and on day 5 for a blastocyst. In some cases, in which the endometrial thickness was <7.0 mm, the period of E2 administration was extended. It has been reported that when the endometrium is <7 mm, the pregnancy rate decreases. 26 However, in some patients, the endometrium does not become thicker even if E2 is increased or prolonged. Thus, embryo transfer was not canceled because of insufficient thickness in any case. Single embryo transfer of a cleavage-stage embryo or blastocyst was performed. If the patient had a positive beta-human chorionic gonadotropin blood test result in pregnancy week 4, fetal growth was assessed weekly thereafter by transvaginal ultrasonography. The administration of vaginal suppositories was continued until pregnancy week 10 or pregnancy termination.
The primary outcome measure of this study was CPR, and the secondary outcome measures were FHR and MR.

| Adverse events
All adverse reactions other than the side effects described on the drug package, such as thrombosis, headache, somnolence, genital bleeding, and diarrhea, were recorded. Patients were able to withdraw from the study if severe adverse events occurred.

| RE SULTS
We evaluated 259 cycles for eligibility, of which 254 were randomized to four study groups. Twenty-four cycles could not be followed up or the patients discontinued the study. Reasons for dropping out of the study were as follows: Per patients' request, participation in five cycles was declined even after judging that there was no problem in the evaluation of eligibility. Two cycles were withdrawn from the Lutinus group because two patients wanted to transfer two embryos. Three cycles were withdrawn from the Utrogestan group because two patients wanted another drug, and one patient had no implantable embryo. Eight cycles were withdrawn from the Luteum group because one patient wanted another drug, four patients wanted to transfer two embryos, two patients had no implantable embryo, and one patient had severe vaginal bleeding and changed the drug. Six cycles were withdrawn from the Crinone group because three patients wanted to transfer two embryos, two patients had no implantable embryo, and one patient wanted another drug. No serious adverse events occurred in the women involved in this study.
As a result, the number of cycles included in the final analysis was 235 (Figure 1). The number of patients who participated in this study was 183; thus, the average participation was 1.2 times. The number of cycles assigned to the same vaginal P4 as the previously assigned cycle was only 10.
The patient demographics are summarized in Table 1. Blastocysts that were assessed as 3BB or higher using the Gardner classification were identified as high-quality embryos. The results showed no significant differences between the study groups regarding the baseline characteristics: age (P = .169), BMI (P = .384), pregnancy history (P = .732), previous transfers (P = .679), endometrial thickness (P = .132), transferred embryo (P = .909), and quality of the blastocyst (P = .384).
Although we consider the median age of the patients (35-37 years) to be higher than that of those in other countries, this age group is commonly dealt with in fertility clinics located in Japan.
Pregnancy outcome data are shown in Table 2. There were no statistically significant differences between the four groups in the rates of CPR (Lutinus, Utrogestan, Luteum, and Crinone: 34.9%, 33.3%, 37.5%, and 35.7%, respectively; P = .976), FHR (26.9%, 31.6%, 30.3%, and 25.0%, respectively; P = .858), and MR (31.8%, 10.0%, 19.0%, and 30.0%, respectively; P = .306). The crude odds ratios (ORs) for these comparisons are shown in Table 3. We did not find any significant difference between the four study groups regarding CPR, FHR, and MR in univariate OR. We performed a multivariate logistic regression analysis to eliminate the confounding effects of age, BMI, and the number of previous transfers (Table 3). We also found that after controlling for confounders, there was no significant difference between the study groups.

| D ISCUSS I ON
This study's results indicated that no significant differences were found between the four preparations in terms of CPR, FHR, and MR. All four of the preparations are the same natural P4 agents, but their daily doses range from 90 mg to 300 mg, 600 mg, and 800 mg.
Regardless of this, the fact that all four have the same pregnancy rates and MRs is extremely interesting. Over the history of vaginal preparations, the form of the preparations has undergone development from vaginal suppositories to gelatin capsules, bioadhesive gel, and most recently, foam agents. [33][34][35][36][37][38][39][40][41] Crinone and Lutinus, which are newer preparations, contain smaller amounts of P4, presumably because they have superior solubility and absorption rates. In addition, Crinone and Lutinus are applied with the use of an applicator. It has been reported that the use of an applicator to place the preparation deeper into the vagina allows it to be efficiently transferred to the endometrium. 42 This is the likely reason why these two preparations require smaller amounts of P4.
The methods of application and doses currently recommended by the manufacturers of all four preparations are indicated for use in Fresh-ET procedures. [43][44][45][46] As a result, it is unknown whether the same doses are effective when the preparations are used for HRT-FET. It has been reported that higher doses are better when used in HRT-FET. One such study reported that the pregnancy rate was higher when the dose of Crinone is double the recommended dose of 90 mg/d (ie, 180 mg/d). 47 Since this was a retrospective study, bias may have been involved. A second such study reported that the pregnancy rate was higher when Utrogestan was used at a dose of 1200 mg/d rather than 900 mg/d. 48 Because additional doses are required when the blood P4 level is <9 ng/mL on day 5 of the luteal phase, this study was not properly conducted. The third such study TA B L E 1 Patient demographics for frozen embryo transfer cycles using four vaginal progesterones for luteal phase support

ACK N OWLED G EM ENTS
None.

CO N FLI C T O F I NTE R E S T
None.