Endocrine Responses to Triptorelin in Healthy Women, Women With Polycystic Ovary Syndrome, and Women With Hypothalamic Amenorrhea

Abstract

Context

Limited data exist regarding whether the endocrine response to the gonadotropin-releasing hormone receptor agonist (GnRHa) triptorelin differs in women with polycystic ovary syndrome (PCOS) compared with healthy women or those with hypothalamic amenorrhea (HA).

Objective

We compared the gonadotropin response to triptorelin in healthy women, women with PCOS, or those with HA without ovarian stimulation, and in women with or without polycystic ovaries undergoing oocyte donation cycles after ovarian stimulation.

Methods

The change in serum gonadotropin levels was determined in (1) a prospective single-blinded placebo-controlled study to determine the endocrine profile of triptorelin (0.2 mg) or saline-placebo in healthy women, women with PCOS, and those with HA, without ovarian stimulation; and (2) a retrospective analysis from a dose-finding randomized controlled trial of triptorelin (0.2-0.4 mg) in oocyte donation cycles after ovarian stimulation.

Results

In Study 1, triptorelin induced an increase in serum luteinizing hormone (LH) of similar amplitude in all women (mean peak LH: healthy, 52.3; PCOS, 46.2; HA, 41.3 IU/L). The AUC of change in serum follicle-stimulating hormone (FSH) was attenuated in women with PCOS compared with healthy women and women with HA (median AUC of change in serum FSH: PCOS, 127.2; healthy, 253.8; HA, 326.7 IU.h/L; P = 0.0005). In Study 2, FSH levels 4 hours after triptorelin were reduced in women with at least one polycystic morphology ovary (n = 60) vs normal morphology ovaries (n = 91) (34.0 vs 42.3 IU/L; P = 0.0003). Serum anti-Müllerian hormone (AMH) was negatively associated with the increase in FSH after triptorelin, both with and without ovarian stimulation.

Conclusion

FSH response to triptorelin was attenuated in women with polycystic ovaries, both with and without ovarian stimulation, and was negatively related to AMH levels.

Gonadotropin-releasing hormone (GnRH) is a decapeptide with a short half-life of 2 to 4 minutes (1). In 1971, modifications of GnRH were created to produce longer acting GnRH receptor agonists (GnRHa), enabling more potent and longer lasting stimulation of the GnRH receptor (2). Conversely, sustained administration can lead to desensitization of the GnRH receptor and downregulation of the reproductive endocrine axis (3). With the advent of competitive GnRH antagonists, it became possible to use GnRHa treatment during in vitro fertilization (IVF) treatment to stimulate the reproductive axis (4). The GnRHa displaces the GnRH antagonist from the GnRH receptor and can induce a gonadotropin rise sufficient to mature oocytes (5). GnRHa can be used in place of human chorionic gonadotropin (hCG) to trigger oocyte maturation in women at high a priori risk, such as women with polycystic ovary syndrome (PCOS), to reduce the risk of ovarian hyperstimulation syndrome (OHSS) (6). Of note, GnRHa induce a rise in follicle-stimulating hormone (FSH) whereas hCG provides stimulation of only the luteinizing hormone (LH) receptor (7); however, the importance of FSH activity for inducing oocyte maturation is still uncertain (8).

GnRHa are the trigger of choice in women at increased risk of OHSS, such as those with PCOS; however, there are only limited data describing whether the endocrine response to triptorelin differ in women with PCOS compared with healthy women or those with hypothalamic amenorrhea (HA). Therefore, we performed a single-blinded placebo-controlled study to compare the endocrine response to triptorelin 0.2 mg in eumenorrheic women as well as in women with PCOS or HA in the absence of ovarian stimulation. Additionally, to understand whether the observed results were also present in women undergoing ovarian stimulation, we analyzed data from a dose-finding randomized controlled trial of triptorelin after ovarian stimulation in oocyte donation cycles (9) to assess whether the endocrine response to triptorelin differed in women with polycystic ovarian morphology compared with women with normal morphology ovaries.

Methods

Study 1

Ethical approval

The study was conducted with approval from the West London Research Ethics committee, London, (reference: 12/LO/0507). The study was part of a larger body of work assessing the endocrine response to a kisspeptin receptor agonist (10). The study was conducted in accordance with the Declaration of Helsinki and all participants provided written informed consent.

Participants

We recruited 9 healthy eumenorrheic women, 6 women with PCOS, and 6 women with HA via advertisements in the local press. Eligibility was confirmed following detailed medical, endocrine, and menstrual history; clinical examination; and blood tests, including an anterior pituitary panel and serum β-hCG level to exclude pregnancy. The inclusion criteria for healthy eumenorrheic women were as follows: age 18 to 35 years, regular menstrual cycles (menstrual cycle length <35 days) and body mass index (BMI) 18 to 30 kg/m², not taking any medications or hormonal contraception. Healthy women did not routinely undergo transvaginal ultrasound, but all women had AMH levels assessed. None of the women in this group had clinical or biochemical hyperandrogenism. All women had hyperprolactinemia excluded with a serum prolactin level. All women had congenital adrenal hyperplasia excluded based on a 17-hydroxy-progesterone levels (17-OHP) < 6 nmol/L. Women with PCOS were defined according to the 2018 international PCOS guideline, namely at least 2 of: (1) history of secondary amenorrhea or oligomenorrhea (defined as ≤8 menstrual cycles per year); (2) polycystic ovarian morphology on ultrasound scan with the presence of at least 20 follicles in at least one ovary; and (3) the presence of hyperandrogenism. Biochemical hyperandrogenism was defined as total testosterone level above the upper limit of the reference range (>2.0 nmol/L). Clinical hyperandrogenism was defined as a score of ≥8 on the Ferriman-Gallwey hirsutism score (11). Women with HA were diagnosed according to the American Endocrine Society guideline (12); specifically, women with HA had secondary amenorrhea of at least 6 months’ duration, in the presence of significant weight loss, vigorous exercise, or emotional stress, with low LH, and normal or low FSH levels.

Study 1 protocol

All study visits were carried out at the Clinical Research Unit, Imperial College Healthcare NHS Trust and commenced in the morning (∼8 Am). Pregnancy was excluded using a urine pregnancy test at the beginning of each study visit. All but one of the women with PCOS had amenorrhea rather than only oligomenorrhea, and a 7-day course of medroxyprogesterone was used to induce a withdrawal bleed (medroxyprogesterone, 10 mg twice daily for 7 days). The study was conducted within days 1 to 4 following a withdrawal bleed in women with PCOS. All women received a subcutaneous bolus of either triptorelin (0.2 mg) or vehicle (0.9% saline) in the early follicular phase of the menstrual cycle (day 1 to day 4) after a spontaneous or induced menstrual bleed. The order of administration (triptorelin or saline) was randomized, and each study visit took place once per menstrual cycle. Women with HA did not receive a course of progesterone to induce a withdrawal bleed, as this was unlikely to be successful, but study visits were scheduled to occur once per month to ensure consistent treatment washout times. Blood sampling for reproductive hormone levels at −30 minutes, 0 minutes (just prior to GnRHa), at 5 minutes after GnRHa, 15 minutes after GnRHa, 30 minutes after GnRHa, and thereafter every 30 minutes for 10 hours, with an additional sample taken at 24 hours following GnRHa. The arrows represent the time of blood samples being taken (Fig. 1).

Figure 1.

Study protocol. Diagram outlining the protocol for Study 1. A subcutaneous bolus of GnRHa (triptorelin 0.2 mg) or 0.9% saline was administered at T = 0 hours, at 5 minutes after GnRHa, 15 minutes after GnRHa, 30 minutes after GnRHa, and thereafter every 30 minutes for 10 hours, with an additional sample taken at 24 hours following GnRHa. Abbreviations: LH, luteinizing hormone; FSH, follicle-stimulating hormone.

Study 1 peptide

GnRH agonist triptorelin (Decapeptyl) was purchased from Clinigen Pharmaceuticals (Germany) and Ipsen Pharmaceuticals (France).

Study 1 hormone assays

Blood samples for measurement of serum LH, FSH, and estradiol were collected in simple Vacutainer tubes (Becton Dickinson, Franklin Labs, New Jersey). Blood samples were centrifuged at least 1 hour after collection at 3000 rpm for 10 minutes. Serum samples generated from the study were frozen at −20 °C. Serum LH, FSH, and estradiol were measured using automated chemiluminescent immunoassays (Abbott Diagnostics, Maidenhead, UK). Interassay coefficients of variation were as follows: LH 2.7%; FSH 3.0%; and estradiol 3.0%. Limits of detection for each assay were as follows: LH 0.05 IU/L, FSH 0.05 IU/L, and estradiol 37 pmol/L. Reference ranges were as follows: LH in IU/L, 2 to 10 (follicular), 20 to 60 (mid cycle), 4 to 14 (luteal); FSH in IU/L, 1.5 to 8.0 (follicular and luteal), 10 to 50 (mid cycle), and estradiol in pmol/L, less than 300 (early follicular), 400 to 1500 (mid cycle), 200 to 1000 (luteal).

Study 2

Endocrine profile after GnRHa triptorelin following ovarian stimulation

In order to verify whether the endocrine profile after triptorelin was similar in the context of ovarian stimulation, we used data from a previously reported randomized controlled trial of triptorelin in oocyte donation stimulated cycles. The study protocol is described in full as published in 2016 by Vuong et al (9). In brief, women recruited in this study were aged 18 to 35 years with BMI < 28 kg/m² and had normal ovarian reserve (AMH >1.25 ng/mL or antral follicle count ≥ 6) (9). Women were excluded if they had PCOS or any other chronic medical condition (9). Women underwent oocyte donation cycles receiving ovarian stimulation with corifollitropin alfa, followed by cotreatment with ganirelix and follitropin-β. Final oocyte maturation was performed with a single subcutaneous bolus of triptorelin at doses of either 0.2, 0.3, or 0.4 mg (10). The study was conducted with approval from the Institutional Review Board (reference number: NCKH/CGRH_01_2014) in accordance with the Declaration of Helsinki.

Statistical methods

Analyses were conducted using Graphpad Prism version 9.0. Normality was determined using the D’Agostino-Pearson test. Parametrically distributed data were presented as mean (±SD), whereas nonparametric data were presented as median (interquartile range [IQR]). Statistical comparison across intervention groups was performed using one-way analysis of variance (ANOVA) with post hoc Tukey's multiple comparison test or Kruskal-Wallis test with post hoc Dunn's test, as appropriate. A P value <0.05 was considered statistically significant. The change in LH was used to account for different baseline levels in the different groups and was calculated as the difference in LH compared with the baseline LH level (measured at Time = 0 just prior to GnRHa administration). The area under the curve (AUC) of this over 24 hours is presented in Fig. 2B, 2D, and 2F. Simple linear regression analysis was used to determine the relationship between peak change in serum LH and serum FSH with AMH and inhibin B levels as shown in Fig. 3. Data in Fig. 4A were analyzed using a mixed-effects model with Sidak's multiple comparisons test. Data in Fig. 4B were analyzed using a Mann-Whitney U test. Figure 4C was analyzed using a Kruskal-Wallis test with post hoc Dunn's multiple comparison test. Data in Fig. 4D and 4E were analyzed using simple linear regression.

Figure 2.

Study 1: reproductive hormone responses following triptorelin and vehicle in healthy women, women with PCOS, and women with HA. A, Mean ± SEM of serum LH (IU/L) after a single subcutaneous bolus of triptorelin (0.2 mg) in healthy women in gray circles (n = 9), women with PCOS in red triangles (n = 6), and women with HA in blue squares (n = 6) and following administration of 0.9% saline (all women) in black inverted triangles. B, Median (IQR) of the area under the curve (AUC) of change in serum LH (IU.h/L) in healthy women in gray circles (n = 9), women with HA in blue squares (n = 6) and women with PCOS in red triangles (n = 6) over 24 hours. Comparison was made with a one-way ANOVA and Tukey's multiple comparisons test. C, Mean ± SEM of serum FSH (IU/L) after a single subcutaneous bolus of triptorelin (0.2 mg) in healthy women in gray circles (n = 9), women with PCOS in red triangles (n = 6), and women with HA in blue squares (n = 6) and following administration of 0.9% saline (all women) in black. D, Median (IQR) of the AUC of change in serum FSH (IU.h/L) in healthy women in gray circles (n = 9), women with HA in blue squares (n = 6), and women with PCOS in red triangles (n = 6) over 24 hours. Comparison was made with a one-way ANOVA and Tukey's multiple comparisons test. P = 0.0193 for healthy vs PCOS, P = 0.0003 for HA vs PCOS. E, Mean ± SEM of serum estradiol (pmol/L) after a single subcutaneous bolus of triptorelin (0.2 mg) in healthy women in gray circles (n = 9), women with PCOS in red triangles (n = 6), and women with HA in blue squares (n = 6) and following administration of 0.9% saline in black inverted triangles. F, Median (IQR) of AUC of change in serum estradiol (pmol.h/L) in healthy women in black (n = 9), women with HA in blue (n = 6), and women with PCOS in red (n = 6) over 24 hours. Comparison was made with a one-way ANOVA and Tukey's multiple comparisons test.

Figure 3.

Study 1: associations between reproductive hormone responses and baseline characteristics following triptorelin in healthy women, women with PCOS, and women with HA. A, Relationship between maximal change in serum FSH (IU/L) and maximal change in serum LH (IU/L) achieved following administration of triptorelin in healthy women in black circles (n = 9), women with HA in blue squares (n = 6), and women with PCOS in red triangles (n = 6). Simple linear regression: r² = 0.28, P = 0.014. B, Median (IQR) of the ratio of maximal change in LH over maximal change in FSH in healthy women in black circles (n = 9), women with HA in blue squares (n = 6), and women with PCOS in red triangles (n = 6). Comparison was made with a one-way ANOVA and Tukey's multiple comparisons test. P = 0.0452 for healthy vs PCOS, P = 0.0092 for HA vs PCOS. C, Relationship between the maximal change in serum LH (IU/L) and baseline serum anti-Müllerian hormone (AMH) levels (pmol/L) in healthy women in black circles (n = 9), women with HA in blue squares (n = 6), and women with PCOS in red triangles (n = 6). D, Relationship between the maximal change in serum FSH (IU/L) and baseline serum AMH levels (pmol/L) in healthy women in black circles (n = 9), women with HA in blue squares (n = 6), and women with PCOS in red trianlges (n = 6). Simple linear regression: r² = 0.51, P = 0.0029. E, Relationship between the maximal change in serum LH (IU/L) and baseline Inhibin B (pg/mL) in healthy women in black circles (n = 9), women with HA in blue squares (n = 6), and women with PCOS in red triangles (n = 6). F, Relationship between the maximal change in serum FSH (IU/L) and baseline inhibin B (pg/mL) in healthy women in black circles (n = 9), women with HA in blue squares (n = 6), and women with PCOS in red triangles (n = 6).

Figure 4.

Study 2: comparison of FSH responses between women with normal ovarian morphology and women with polycystic ovarian morphology following triptorelin administration in a stimulated ovarian cycle. A, Mean ± SD of serum FSH (IU/L) over time in women with normal ovarian morphology (n = 101) and women with polycystic ovarian morphology (n = 64) following triptorelin administration in a stimulated ovarian cycle. Comparison was made with a mixed-effects model and Sidak's multiple comparisons test. P ≤ 0.0001. B, Scatterplot of median (IQR) of serum FSH (IU/L) in women with normal ovarian morphology (Norm Norm) (n = 91) and women with one or more polycystic ovary (≥1 PCOM ovary) (n = 60) 4 hours after triptorelin administration in a stimulated ovarian cycle. Comparison was made with a Mann-Whitney test; P = 0.0003. C, Scatterplot of median (IQR) of serum FSH (IU/L) in women with normal ovarian morphology (Norm Norm) (n = 91), women with 1 polycystic ovary (Norm PCOM) (n = 29), and women with 2 polycystic ovaries (PCOM PCOM) (n = 31) 4 hours after triptorelin administration in a stimulated ovarian cycle. Comparison was made with a Kruskal-Wallis test with Dunn's multiple comparisons; P = 0.0001 for Norm Norm vs PCOM PCOM. D, Relationship between serum FSH (IU/L) levels 4 hours following triptorelin administration and total antral follicle count (AFC) (n = 151) in a stimulated ovarian cycle. Simple linear regression: r² = 0.10, P ≤ 0.0001. E, Relationship between serum FSH (IU/L) levels 4 hours following triptorelin administration and baseline serum anti-Müllerian hormone (AMH) levels (pmol/L) (n = 151) in a stimulated ovarian cycle. Simple linear regression: r² = 0.19, P ≤ 0.0001.

Results

Study 1

Baseline characteristics

Following screening, 9 healthy eumenorrheic women, 6 with PCOS, and 6 with HA were recruited to the study. Baseline characteristics and baseline reproductive hormone profiles of the study participants are presented in Table 1. There was no significant difference in age, weight, BMI, baseline FSH, estradiol, or sex hormone binding globulin (SHBG) levels (Table 1) between the 3 groups. Serum LH and serum anti-Müllerian hormone (AMH) levels were significantly higher in women with PCOS (P = 0.003 for LH, and P = 0.005 for AMH).

Table 1.

Baseline characteristics of healthy women, women with PCOS, and women with HA

CLINICAL CHARACTERISTICS	HEALTHY WOMEN (N = 9)	WOMEN WITH PCOS (N = 6)	WOMEN WITH HA (N = 6)	P VALUE
AGE, YEARS	26.0 (21.0, 32.8)	25.5 (23.5, 28)	25.0 (23.0, 30.8)	0.72
MASS, KG	64.4 (55.9, 71.2)	67.1 (52.4, 70.6)	56.4 (51.5, 62.3)	0.21
BMI, KG/MCSSUPSTART2CSSUPEND	24.5 (20.1, 25.5)	24.4 (21.2, 27.7)	20.7 (19.4, 23.1)	0.12
SERUM LH, IU/L	3.7 (3.0, 4.4)	8.2 (5.7, 13.9)	2.9 (1.4, 3.6)	0.003
SERUM FSH, IU/L	5.1 (2.9, 5.9)	5.4 (2.9, 5.7)	5.2 (4.6, 6.3)	0.18
SERUM ESTRADIOL, PMOL/L	93.0 (72.3, 135.3)	121.0 (68.5, 202.5)	72.5 (50.5, 110.3)	0.49
SHBG, NMOL/L	69.0 (48.8, 109.5)	39.0 (32.0, 48.3)	65.0 (39.8, 100.3)	0.08
SERUM AMH, PMOL/L	23.9 (2.4, 36.8)	65.8 (29.9, 98.3)	20.8 (14.9, 23.2)	0.005

Median (interquartile centile) of baseline demographic characteristics and reproductive hormone profiles for each group. Kruskal-Wallis test with Dunn's multiple comparison was used for comparisons between the different groups. LH was significantly lower in women with HA compared with PCOS (P = 0.005). AMH was significantly higher in women with PCOS compared to women with HA (P = 0.021) or to women with HA (P = 0.024). Abbreviations: AMH, anti-Müllerian hormone; BMI, body mass index; FSH, follicle-stimulating hormone; LH, luteinizing hormone; SHBG, sex hormone binding globulin.

Maximum level of gonadotropin change after triptorelin in healthy women

In healthy women during the follicular phase, serum LH levels reached a mean maximum of 49.0 IU/L at approximately 4 hours after triptorelin administration and remained elevated with a mean of 28.0 IU/L at 24 hours following administration (Fig. 2A). There was no correlation between the LH level after triptorelin and BMI (P = 0.91). Serum FSH peaked at 5 hours after triptorelin to a mean of 20.3 IU/L and levels remained elevated with a mean of 13.2 IU/L at 24 hours following administration (Fig. 2C). Serum estradiol rose gradually, reaching a mean level of 566 pmol/L at 10 hours post administration and remained elevated with a mean estradiol level of 561 pmol/L at 24 hours (Fig. 2E).

Gonadotropin changes after triptorelin in women with PCOS and HA

The change in gonadotropins appeared to be expedited in women with HA in comparison to healthy women or those with PCOS, with a greater LH level at 1 hour post triptorelin than in women with PCOS or in eumenorrheic women (mean change at serum LH at 1 hour: 8.8 IU/L in healthy women, 29.1 IU/L in women with HA and 12.7 IU/L in women with PCOS, P = 0.0012 for HA vs healthy and P = 0.0138 for HA vs PCOS) (Fig. 2A). However, the level at 4 hours was greater in healthy women and those with PCOS than women with HA (mean change at serum LH at 4 hours 45.7 IU/L in healthy women, 31.3 IU/L in women with HA, and 36.9 IU/L in women with PCOS), such that the area under the curve (AUC) of change in serum LH following triptorelin administration over 10 hours did not differ between the groups (Fig. 2B).

The change in serum FSH was markedly attenuated in women with PCOS in comparison with both healthy women and those with HA (Fig. 2C). The AUC of change in serum FSH following triptorelin administration was significantly reduced in women with PCOS (P = 0.019 for healthy vs PCOS; P = 0.0003 for HA vs PCOS) (Fig. 2D). Despite the reduced change in serum FSH, the change in estradiol levels (Fig. 2E) and the AUC of change in serum estradiol following triptorelin administration (Fig. 2F) did not differ between the 3 groups.

FSH response after triptorelin negatively correlates with AMH level

There was a positive association between the maximal stimulated LH and the maximal stimulated FSH following triptorelin (r² = 0.28, P = 0.014) (Fig. 3A), albeit women with HA tended to have higher FSH whereas women with PCOS tended to have lower FSH for similar change in serum LH (Fig. 3A). Indeed, the PCOS group had the highest ratio of change in serum LH to change in serum FSH, compared with both the HA and healthy groups (P = 0.045 for healthy vs PCOS; P = 0.0092 for HA vs PCOS) (Fig. 3B). The maximal change in FSH (Fig. 3D), but not the maximal change in serum LH (Fig. 3C), negatively correlated with serum AMH levels (r² = 0.51, P = 0.0029). Furthermore, baseline inhibin B levels were not associated with the maximal change in LH or the maximal change in FSH (Fig. 3E and 3F).

Study 2

Endocrine response to triptorelin in the context of ovarian stimulation in oocyte donation cycles

In order to investigate whether women with polycystic ovaries also had lower change in serum FSH after triptorelin after ovarian stimulation, we analyzed data from a randomized controlled trial of triptorelin at doses 0.2 to 0.4 mg in oocyte donation cycles (14). Concordantly, women with at least one polycystic ovarian morphology (PCOM) ovary (n = 60), had a lower maximal change in serum FSH at 4 hours than women with normal morphology ovaries (n = 91) (−8.4; 95% CI, −11.7 to −5.2; P < 0.0001) (Fig. 4A). Mean FSH levels at 4 hours following triptorelin in women with at least one PCOM ovary were significantly lower compared to women with normal ovarian morphology (34.1 vs 42.3 IU/L; P = 0.0003) (Fig. 4B). There was no difference in the mean dose of triptorelin received by those without PCOM ovaries (0.30 mg) compared with those with 1 PCOM ovary (0.30 mg), or in those with 2 PCOM morphology ovaries (0.29 mg). Women in these 3 categories had similar age (27.8, 25.0, and 26.6 yrs; P = 0.001), bodyweight (51.2, 49.2, 53.9 kg/m²; P = 0.04) and BMI (20.8, 19.9, 21.7 kg/m²; P = 0.03). The mean ± SD of FSH at 4 hours after triptorelin was 42.3 ± 14.4 IU/L in women with 2 normal morphology ovaries, 37.5 ± 9.9 IU/L in women with only one PCOM ovary, and 30.9 ± 9.2 IU/L in women with 2 PCOM ovaries (Fig. 4C). The FSH level at 4 hours after triptorelin was negatively correlated with total antral follicle count (r² = 0.10, P < 0.0001) (Fig. 4D) and AMH level (r² = 0.19, P < 0.0001) (Fig. 4E).

Discussion

In the present study, we compared the hormonal response to the GnRHa triptorelin in healthy eumenorrheic women, women with PCOS, and those with HA. Women from all 3 groups had a similar maximal change in serum LH, however women with HA exhibited an earlier rise with higher LH levels observed at ∼1 hour after administration rather than at ∼4 hours in healthy women and those with PCOS. Notably, women with PCOS exhibited an attenuated change in serum FSH compared with healthy eumenorrheic women and women with HA. However, despite the attenuated change in FSH, women with PCOS had a comparable increase in serum estradiol levels. Furthermore, the maximal change in serum FSH negatively correlated with baseline AMH levels.

Several GnRHa drugs, including triptorelin, naferelin, leuprolide, and buserelin, are in clinical use but there are only limited data describing the endocrine profile induced. The relative potency (EC₅₀) in vitro with respect to IP₃ accumulation in COS-1 cells transfected with the human GnRH I receptor is as follows: endogenous GnRH 0.054 nM (95% CI, 0.015-0.19), buserelin 0.031 nM (95% CI, 0.014-0.069), and triptorelin 0.019 nM (95% CI, 0.009-0.042) (13). The half-lives of the different agonists are as follows: endogenous GnRH 2 to 4 minutes; triptorelin 4 hours (14); naferelin 3 to 4 hours (15); leuprolide 3.6 hours (16); and buserelin 1.3 hours (17). Thus, triptorelin has a longer half-life than buserelin or native GnRH, and subcutaneous injection may increase the half-life of triptorelin by 10-fold (18).

A study has compared different GnRHas at different doses in an ovulation induction protocol with no clear differences in efficacy observed (19). Leuprolide acetate has been used to trigger oocyte maturation during IVF treatment at doses ranging from 0.5 to 4 mg (20-22). Buserelin is most commonly used at 0.5 mg (23). Triptorelin is most commonly used at 0.2 mg; however, doses from 0.1 mg to 0.4 mg have been used (9, 24). A large randomized controlled trial conducted in a Vietnamese population demonstrated no differences in the gonadotropin rise or in the efficacy of inducing oocyte maturation between doses of triptorelin of 0.2 mg, 0.3 mg, or 0.4 mg, suggesting that 0.2 mg is near the upper end of the dose-response range (9). More recently, another study showed that a lower dose of 0.1 mg of triptorelin induced comparable gonadotropin rises and oocyte maturation rates when compared with higher doses of 0.2 and 0.4 mg (25). Taken together, the studies support the notion that doses of 0.1 to 0.4 mg of triptorelin result in comparable outcomes in terms of oocyte maturation.

We observed similar peak and AUC of change in serum LH following triptorelin administration between the 3 groups. This is consistent with a study reporting that serum LH changes after 50 μg of native GnRH were similar in women with HA, anorexia nervosa, and in those with secondary “stress” amenorrhea as in healthy women (26). The maximal change in serum LH in the present study is ∼38 IU/L, whereas maximal change in serum LH in a stimulated cycle are typically approximately 3 to 5 times greater, albeit with similar timing of maximal change in serum LH levels (9, 25). Vuong et al reported a maximal serum LH level of 140 IU/L at 4 hours after triptorelin 0.2 mg in a stimulated oocyte donor cycles (9). The amplitude of change in serum LH after GnRHa in the setting of ovarian stimulation (140 IU/L) is higher than that occurring during the physiological LH surge (∼52 IU/L) (27). Similarly, we observed an LH level in all 3 groups of ∼31 IU/L at 8 hours in unstimulated cycles in the present study, whereas Lainas et al reported a 3-fold greater LH level of ∼99 IU/L at 8 hours post triptorelin 0.2 mg in a stimulated cycle (25). A key difference between the stimulated and unstimulated cycle is the baseline estradiol level, which exceeds the threshold for positive feedback in the stimulated cycle; estradiol levels in the present study were in the region of ∼100 pmol/L at baseline, whereas as in the stimulated cycles discussed, preinjection estradiol levels were >7000 pmol/L (9, 25).

Interestingly, in stimulated cycles, LH levels fall more rapidly after peaking at 4 hours post triptorelin, than in the unstimulated cycle reported here (9). In the present study, LH levels at 10 hours were 23.6% lower than peak LH level at 4 hours, whereas in the stimulated cycle, LH levels at 12 hours had fallen by 77% (9). A possible explanation is that this is due to estradiol levels being below the threshold for estradiol-induced positive feedback at baseline, but that they exceed this threshold after triptorelin, such that a further LH rise could have been induced. In contrast, in the stimulated cycle, baseline estradiol levels already exceed the threshold for positive feedback even prior to administration of triptorelin. An alternative possibility is that in the stimulated cycle, multi-follicular growth may result in the release of other inhibitory factors such as gonadotropin surge attenuating factor (GnSAF), which could prevent the sustained rise in LH levels (28).

Basal LH levels correlate with stimulated LH levels following GnRHa (29). In women undergoing an oocyte donation cycle with ovarian stimulation using a GnRHa to induce oocyte maturation, the baseline LH level just prior to GnRHa administration was associated with a greater rise in LH at 4 hours after GnRHa (27). While pre-GnRHa LH levels are known to be associated with a higher LH level after triptorelin in stimulated cycles (27), women with PCOS (who are known to have higher baseline LH levels) did not appear to have an increased LH response after GnRHa in comparison with other groups in the present study. Previous studies have suggested an increased change in serum LH to GnRHa administration in women with PCOS; absolute mean change in serum LH after 0.10 µg/kg of the GnRH agonist ([(imBzl) D-His6, Pro9-NEt]-GnRH (D-His) resulted in an increase in LH from 12.64 IU/L to 172 IU/L at 4.2 hours in 6 women with PCOS, compared to a change in serum LH from 3.96 IU/L to 65 IU/L in 6 healthy eumenorrheic women (30). Similarly, absolute change in serum LH following GnRH stimulation (2-20 μg) were 2- to 3-fold greater in PCOS (BMI 34.7 kg/m²; n = 13) than in healthy women (BMI 26.8 kg/m², n = 13) (31). The maximal change in serum LH after GnRH correlated positively with basal LH values but negatively with BMI (31). Accordingly, due to high basal LH values in PCOS (PCOS women, 7.5 ± 1.2 vs eumenorrheic women, 3.6 ± 0.4 IU/L), the percentage rise in serum LH was not increased (31). Consequently, absolute increases in serum LH to pituitary stimulation may not necessarily indicate inherent heightened pituitary sensitivity but rather reflect increased basal LH levels. Our findings are consistent with that of Hirshfeld-Cytron and colleagues who demonstrated similar LH rises in women with PCOS compared to controls but showed reduced increases in FSH (PCOS ∼18 vs controls 32 IU/L)(32).

There is also a suggestion that the rise in LH after GnRHa was expedited in women with HA, which is consistent with data on the rise in LH after a kisspeptin receptor agonist in women with HA, which was ascribed to increased hypothalamic GnRH neuronal kisspeptin receptors (33); however, the explanation for an expedited rise after GnRHa (although not as stark as following kisspeptin receptor agonist) has yet to be fully elucidated. Women with PCOS required a seven-day course of progesterone to induce a withdrawal bleed prior to each study visit. However, women with HA and healthy women with eumenorrheic cycles did not receive progesterone administration. Consequently, it is possible that progesterone administration in the previous cycle could have negatively impacted basal LH levels and thus the LH response to triptorelin (34).

A notable finding was the lower change in serum FSH following GnRHa in women with PCOS in comparison to women with HA and healthy women. Indeed, the change in serum FSH at 4 hours following triptorelin in women with PCOS was half of that observed in women with HA and healthy women. There are several reasons that may account for the differences in LH and FSH secretion in women with PCOS. Firstly, GnRH pulse frequency is increased in women with PCOS and synthesis of LH is favored at faster GnRH pulse frequency, whereas FSH secretion is preferred in low GnRH pulse frequency states (35). Moreover, at high GnRH pulse frequency, the GnRH receptor preferentially couples to Gαq_/11 subunit and recruits a second GnRH transduction pathway that inhibits FSH expression and secretion (35). Secondly, FSH release from the pituitary gland is regulated by activin and follistatin aside from alterations in GnRH pulse frequency alone as in the case of LH (36, 37). Increased GnRH pulse frequency increases follistatin, which suppresses activin-induced synthesis and stimulation of FSH (36, 37). In our data, we observed that the AUC of change in serum estradiol following triptorelin was comparable between healthy women, women with PCOS, and women with HA despite the observed differences in FSH, perhaps reflective of a greater number of follicles in women with PCOS.

The finding of attenuated FSH responses following triptorelin in women with PCOS could have relevant clinical impact. Lamb et al conducted a randomized controlled trial demonstrating that a bolus of FSH at the time of hCG administration resulted in a small increase in fertilization rates and an increased oocyte retrieval rate (38). A study by Egbase et al in 2011 suggested that co-administration of FSH on the day of trigger could improve clinical outcomes and reduce the risk of OHSS (39). If that were the case, then the finding of a low FSH response in women with PCOS who are at increased risk of OHSS would be particularly pertinent. However, more recent randomized controlled trials have failed to replicate this finding (8). A retrospective cohort study of women undergoing GnRH antagonist IVF cycles found that among cycles matched for similar estradiol levels, supplementation with FSH on the day of trigger led to a modest improvement in the number of oocytes, blastocysts, and euploid embryos (40). Overall, the benefit of exogenous FSH administration on the day of trigger has not been shown to be significant and has not been investigated in the setting of GnRHa-triggered cycles; however, the reduced FSH response in PCOS would suggest that this could be an important subgroup to consider for dedicated study in the future.

AMH restrains the actions of FSH on both recruitment of follicles into the growing follicle pool and aromatization of androgens to estrogen (41). Women with PCOS had higher AMH levels in comparison with healthy women and women with HA (Table 1). A negative association was observed between serum AMH levels and the maximal change in serum FSH in our study (r = −0.47). Indeed, it is hypothesized that high AMH levels in women with PCOS could be involved in the follicular arrest observed in this condition facilitated by a negative interaction with FSH at the time of selection (42). Furthermore, we observed that the change in serum FSH after triptorelin is also lower in women with polycystic ovarian morphology in comparison to those with normal morphology ovaries in oocyte donation cycles, suggesting that the differential endocrine response to triptorelin also occurs in the context of ovarian stimulation.

AMH is reported to have a direct action on pituitary gonadotropes. Incubation with AMH led to increased expression of the gonadotropin subunits FSHβ- (but not α- and LHβ-) in LβT2 cells (murine clonal gonadotrope-derived cell-line) (43). Incubation of LβT2 cells with AMH and GnRHa enhanced the effect of GnRH on the Fshb gene promoter and synergized with the GnRHa to stimulate Lhb gene promoter activity (44). Moreover, although AMH does not affect GnRH receptor transcription levels (45), GnRH upregulates Amhr2 in gonadotrope cells, especially when GnRH is perfused at high frequency (46). Furthermore, there appear to be reciprocal relationships between GnRH and AMH, as a bolus of GnRH reduced AMH in 31 prepubertal children (47). However, in keeping with our present findings in women, similar results are observed in heifers; although AMH stimulated basal FSH secretion in the anterior pituitary of heifers, it inhibited GnRH-induced FSH secretion (48).

In summary, we evaluated the endocrine profile following triptorelin administration in healthy eumenorrheic women, and women with PCOS or HA. We observed an expedited LH rise in women with HA, and a blunted FSH rise in women with PCOS. Despite the blunted FSH rise, the rise in estradiol was similar in women with PCOS in comparison to other women. Furthermore, blunting of the rise in FSH was correlated with AMH levels but not inhibin B levels. These findings further explicate our understanding of the endocrine changes during pituitary stimulation and suggest that the endocrine response can diverge in different patient groups.

Abbreviations

AMH

anti-Müllerian hormone

ANOVA

analysis of variance

AUC

area under the curve

BMI

body mass index

FSH

follicle-stimulating hormone

GnRH

gonadotropin-releasing hormone

GnRHa

gonadotropin-releasing hormone agonist

HA

hypothalamic amenorrhea

hCG

human chorionic gonadotropin

IQR

interquartile range

IVF

in vitro fertilization

LH

luteinizing hormone

OHSS

ovarian hyperstimulation syndrome

PCOM

polycystic ovary morphology

PCOS

polycystic ovary syndrome

Author Contributions

All authors provided contributions to study conception and design, acquisition of data or analysis and interpretation of data, drafting the article or revising it critically for important intellectual content, and final approval of the version to be published. The most important contributions of each author are: A.A., A.N.C., and W.S.D. designed the study; A.A., M.P., P.C.E., S.A.C., T.D.P., T.M.H., E.G.M., K.P., and L.N.V. conducted experiments and acquired the data; A.A., M.P., P.C.E., S.A.C., T.D.P., and T.M.H. recruited patients for the clinical studies; A.A., M.P., P.C.E., and K.Y.N. analyzed the data; A.A., M.P., P.C.E., E.G.M., and T.H. conducted assays for biochemical analytes; and all authors contributed to the writing and revision of the manuscript. W.S.D. takes final responsibility for this article.

Disclosures

The research was conducted in the absence of any personal, professional, commercial, or financial relationships that could be construed as a potential conflict of interest.

Data Availability

Data will be made available upon reasonable request to the corresponding author.