Therapeutic effects of an oral gonadotropin‐releasing hormone receptor antagonist, relugolix, on preventing premature ovulation in mild ovarian stimulation for IVF

Abstract Purpose Can relugolix, a novel oral gonadotropin‐releasing hormone receptor (GnRH) antagonist, function as an alternative ovulation inhibitor to GnRH antagonist injections? Methods This single‐center, cross‐sectional retrospective study compared premature ovulation rates and clinical outcomes in IVF treatment after mild ovarian stimulation with 40 mg of relugolix (relugolix group) or 0.25‐mg injections of ganirelix acetate or cetrorelix acetate (injection group) between March 2019 and January 2020. Of 247 infertile women (256 IVF cycles) aged ≤42 years, 223 women (230 cycles) were evaluated. In the relugolix and injection groups, we compared 104 and 85 cycles after GnRH antagonist use before the LH surge (LH levels <10 mIU/ml) and 22 and 19 cycles during the LH surge (LH levels ≥10 mIU/ml), respectively. Results Before the LH surge, the ovulation rates in the two groups were very low (p = 0.838), however; during the LH surge, the cycles using relugolix had a high ovulation rate of 40.9% compared with no ovulation in the injection group (p = 0.002). There were no significant differences in embryo culture findings and pregnancy outcomes between the two groups. Conclusions Although relugolix had a high ovulation suppressive effect, when the LH surge occurred, its effect was insufficient.

of 0.25 mg, Merck Biopharma Co., Ltd.) (injection group) between March 2019 and January 2020 in Sugiyama Clinic Shinjuku. Of the 256 IVF cycles in 247 infertile women aged 42 years or younger, we excluded 18 women (18 cycles) with elevated basal luteinizing hormone (LH) levels including 12 with premature ovarian insufficiency and six with polycystic ovary syndrome (PCOS); we also excluded six women (eight cycles) with factors that can influence the outcomes of IVF treatments including five with ovarian endometriomas and one with a husband with azoospermia ( Figure 1). Therefore, 223 women (230 cycles) were evaluated. The incidence of premature LH surge was defined as 10 mIU/ml or more, and the relugolix and injection groups were compared separately by GnRH antagonist administration before and during LH surge. In the relugolix and injection groups, we compared 104 cycles (101 women) and 85 cycles (81 women) when the follicles were sufficiently developed after ovarian stimulation with GnRH antagonist use before LH surge initiation (LH levels <10 mIU/ml) and 22 cycles (22 women) and 19 cycles (19 women) during the LH surge (LH levels ≥10 mIU/ml), respectively.
To confirm the findings on fertilization, embryo culture, and pregnancy outcomes after using relugolix, we evaluated 204 of the 230 IVF cycles after excluding 13 cycles with premature ovulation before oocyte retrieval, four cycles without mature eggs after oocyte retrieval and nine cycles with split conventional IVF and intracytoplasmic sperm injection (ICSI) as a fertilization method ( Figure 2).
We compared the findings of fertilization and embryo culture in 106 IVF cycles (106 women) and 98 IVF cycles (94 women) after administering oral relugolix and ganirelix or cetrorelix injections, respectively. Furthermore, pregnancy outcomes were compared between 132 embryo transfers (ET) cycles (98 women) and 103 ET cycles (78 women) in patients who received oral medicine and GnRH antagonist injections, respectively. This study was approved by the local ethics committee of Sugiyama Clinic, Tokyo, Japan (No. 20-005).

| Mild ovarian stimulation using GnRH antagonists and subsequent IVF-ET
The ovarian stimulation, oocyte retrieval, and IVF-ET treatment methods have been previously described. 11,12 Briefly, the women were administered 50-100 mg of clomiphene citrate (Clomid ® ; Fuji Pharma, Tokyo, Japan) once or twice daily for 5-10 days or 2.5 mg of letrozole (Kobayashi Kako Co., Ltd., Fukui, Japan) daily for 5 days beginning on day 3 of the menstrual cycle in combination with the in- Or, on menstrual day 9-11, when the dominant follicles had reached 17 mm or larger during the LH surge, 40 mg of an oral relugolix tablet or 0.25 mg of a GnRH antagonist injection was administered on the day of trigger administration for oocyte maturation with the use of hCG. Having food diminishes the bioavailability of relugolix 9 ; therefore, we recommended taking relugolix before meals. NSAIDs were not used for the purpose of suppressing ovulation in all cycles. Oocyte retrieval was performed F I G U R E 1 Flowchart of IVF cycles for analyzing the findings of oocyte retrieval. We compared IVF treatment cycles after mild ovarian stimulation using 40 mg of oral relugolix (relugolix group) and 0.25-mg injections of ganirelix acetate or cetrorelix acetate (injection group) between March 2019 and January 2020. Of 256 IVF cycles in 247 infertile women aged ≤42 years, we evaluated 223 women (230 cycles) after excluding 18 women (18 cycles) with elevated basal LH levels including 12 with premature ovarian insufficiency and six with polycystic ovary syndrome; we also excluded six women (eight cycles) with factors influencing the outcomes of IVF treatments including five with ovarian endometriomas and one with a husband with azoospermia after testicular sperm extraction. In the relugolix and injection groups, we compared 104 and 85 cycles after GnRH antagonist use before the start of the LH surge (LH levels <10 mIU/ml) and 22 and 19 cycles during the LH surge (LH levels ≥10 mIU/ml), respectively F I G U R E 2 Flowchart of IVF cycles for the analysis of embryo culture findings and pregnancy outcomes. Of 230 IVF cycles, we evaluated 204 cycles after excluding 13 with premature ovulation, four without mature eggs after oocyte retrieval and nine with split conventional IVF and intracytoplasmic sperm injection (ICSI) as a fertilization method. We compared the findings of fertilization and embryo culture in 106 IVF cycles (106 women) and 98 IVF cycles (94 women) in patients who received oral relugolix or an injection of ganirelix or cetrorelix, respectively. Furthermore, pregnancy outcomes were compared between 132 ET cycles (98 women) and 103 ET cycles (78 women) in patients who received oral medicine and GnRH antagonist injections, respectively transvaginally at 35 h after trigger administration. Ovulation was defined as the rupture of one or more follicles prior to oocyte retrieval.
Either conventional IVF or ICSI was chosen based on the sperm findings and previous fertilization rates. Oocyte maturation was determined by the appearance of the first polar body. Fertilization was confirmed by the detection of two pronuclei and the second polar body. All embryos were evaluated by the Veeck classification at the cleavage stage and the Gardner classification at the blastocyst stage. Morphologically good blastocysts were defined as 5-or 6-day blastocysts of stage 4 or more and grade A or B in both the inner cell mass and the trophectoderm of the Gardner classification.
In our protocol, fresh ET using a cleavage stage embryo and freezing excess blastocysts are mainly performed after ovarian stimulation with letrozole and FSH or hMG injections in the first oocyte retrieval cycle. In the second or later cycles, we performed all blastocysts were cryopreserved using the vitrification method after clomiphene and FSH or hMG and subsequently single blastocyst transfer, because the live birth rate after frozen ET is significantly higher than that in fresh ET. 13 Fresh ET was performed using a soft catheter (Kitazato ET catheter; Kitazato Corporation, Shizuoka, Japan) guided by a transvaginal ultrasound 2, 3, or 5 days after conventional IVF or ICSI. As luteal support, 30 mg of oral dydrogesterone (Dufaston ® Tablets 5 mg; Mylan EPD LLC., Tokyo, Japan) was administered three times daily for 11 days from the day of ET, and 125 mg of hydroxyprogesterone caproate (Progeston ® depot intramuscular injection; Fuji Pharma) was injected on the day of ET and 5 days after ET. In the vitrified-warmed ET cycles, the endometrium was prepared via either spontaneous ovulatory or hormone replacement cycles. In the spontaneous ovulatory cycle, the day of ET was defined as 2, 3, or 5 days after ovulation depending on the development stage of the frozen embryos. The same luteal support using dydrogesterone and hydroxyprogesterone caproate as that for the fresh ET cycle was used. In the hormone replacement cycle, 1.25 mg of conjugated estrogen tablets administered daily (Premarin ® 0.625 mg; Wyeth., Tokyo, Japan) and 2.88 mg of a transdermal estradiol patch

| Statistical analysis
The statistical analyses were performed using GraphPad Prism 5 (GraphPad Software Inc.). Differences between two groups were analyzed using the Mann-Whitney U test or Fisher's exact test, as appropriate. The level of significance was defined as a p value of <0.05.

| Use of GnRH antagonist agents before the start of the LH surge
To identify the therapeutic effect of relugolix for suppressing premature ovulation, we conducted comparisons between 104 cycles using relugolix (relugolix group) and 85 cycles with conventional GnRH antagonist injection (injection group) before the LH surge (LH levels <10 mIU/ml). The characteristics of the cycles are shown in Table 1. Although more women had ovulation disorders in the relugolix group than in the injection group, there were no significant differences in patients' age, pregnancy history, other causes of infertility, previous history of ET cycles, AMH levels, ovarian stimulation with and without clomiphene citrate, oocyte maturation procedure, or hormone data at the day of trigger administration. In our clinic, the costs of 40 mg of relugolix and a 0.25-mg injection of a GnRH antagonist were JPY1,000 and 8,000 (USD9.1 and 72.9), respectively. Although there was no significant difference in the number of doses of GnRH antagonist tablet or injection, the cost for ovulation suppression was significantly lower in the relugolix group than in the injection group (USD13.7 ± 7.2 and 120.1 ± 66.8, respectively, p < 0.001). The ovulation rates in the relugolix and injection groups were very low (1.9% and 2.4%, respectively) and did not differ significantly (p = 0.838). In the oocyte retrieval cycles without premature ovulation, there were no significant differences in the number of retrieved oocytes and their oocyte maturation rates. No women complained of side effects after relugolix use.

| Use of GnRH antagonist agents during the start of the LH surge
When the LH surge was confirmed (LH levels ≥10 mIU/ml), premature ovulation was suppressed using relugolix (relugolix group) in 22 cycles and using GnRH antagonist injection (injection group) in 19 cycles at the day of trigger administration. As shown in Table 2, there were no significant differences in characteristics between the two groups. The cost for ovulation suppression using relugolix was significantly cheaper compared with that for the GnRH antagonist injection (USD9.1 ± 0 and 72.9 ± 0, respectively, p < 0.001). Surprisingly, the cycles using relugolix had a high ovulation rate of 40.9% compared with no cancellation of oocyte retrieval cycles in the injection group (p = 0.002). Of the 11 cycles after ovarian stimulation with clomiphene citrate, premature ovulation occurred with high frequency (six cycles, 54.5%). However, after oocyte retrieval, there were no significant differences in the findings of retrieved oocytes.
No women experienced side effects with relugolix use.

| Clinical outcomes of fertilization, embryo culture, and subsequent pregnancy
To confirm the effect of relugolix on retrieved oocytes, we also examined fertilization and subsequent embryo culture outcomes in 106 cycles in the relugolix group and 98 cycles in the injection group (Table 3). The fertilization rates after conventional IVF and ICSI did not differ between the two both groups (p = 0.511 and 0.560, respectively). In 145 cycles with performing blastocyst cultures, there were no significant differences in the formation rates of blastocysts and morphologically good blastocysts in the relugolix and injection groups (p = 0.882 and 0.401, respectively).
The clinical pregnancy rates in the relugolix and injection groups  (Table 4). There were no significant differences in clinical pregnancy rates after ET cycles with either cleavage or blastocyst stage embryos (p = 0.392 and 0.596, respectively). The miscarriage rates also did not differ significantly between the two groups. This is the first study to examine the applicability of a novel oral GnRH antagonist agent, relugolix, as an ovulation inhibitor in IVF treatment. Oral relugolix (TAK-385) can antagonize gonadotropin secretion from the pituitary in both women and men; therefore, it has been reported as a therapeutic medicine for uterine fibroids, endometriosis, and prostate cancer. 9,10,[14][15][16] Our study also showed relugolix has a high ovulation suppressive effect prior to the LH surge compared with injection preparations of GnRH antagonists.

TA B L E 1 Characteristics and ovulation rates in oral relugolix and GnRH antagonist injection before LH surge
However, 40% or more of the cycles that used relugolix after the LH surge started had ovulation before oocyte retrieval.  groups. In addition, of nine women who had premature ovulation after relugolix intake during LH surge, six women used clomiphene; thus, we included the cycles with clomiphene use. Fourth, we could not confirm the number of the cycles in the patients having meal before relugolix intake during LH surge.
In conclusion, relugolix had a high ovulation suppressive effect and no adverse effects on retrieved oocytes and subsequent pregnancy outcomes following ET in our study. Hence, the ovarian stimulation protocols using oral relugolix can be an alternative stimulation method with low costs and minimum invasion. However, when the LH surge was recognized in mild stimulation protocols, its effect was insufficient to prevent premature ovulation; thus, relugolix should be avoided in such cases. Additional clinical studies including prospective controlled trials in GnRH antagonist cycles are required to confirm the therapeutic effect of relugolix on premature ovulation and pregnancy outcomes in IVF treatment.

ACK N OWLED G M ENTS
We thank Enago (www.enago.jp) for the English language review.

Keiji Kuroda was supported by JSPS KAKENHI Grant Number
18K09273.