Dose-finding study of daily gonadotropin-releasing hormone (GnRH) antagonist for the prevention of premature luteinizing hormone surges in IVF/ICSI patients: antide and hormone levels

Judith A.F. Huirne, Andre C.D. van Loenen, Roel Schats, Joseph McDonnell, Peter G.A. Hompes, Joop Schoemaker, Roy Homburg and Cornelis B. Lambalk1

Division of Reproductive Medicine, Department of Obstetrics and Gynaecology, Vrije Universiteit medical centre (VUmc), Amsterdam, The Netherlands

1 To whom correspondence should be addressed at: Department of Obstetrics and Gynaecology, VUmc, PO Box 7075, 1007 MB Amsterdam, The Netherlands. Email: cb.lambalk{at}vumc.nl


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
BACKGROUND: The aim of this study was to define the minimal effective dose of antide (Iturelix) to prevent premature luteinizing hormone (LH) surges in in vitro fertilization (IVF) patients. METHODS: In a prospective, single centre study, 144 IVF/ICSI patients were stimulated with r-hFSH from cycle day 2 and from cycle day 6 onwards, cotreated with daily 2 mg/2 ml (n=30), 1 mg/ml (n=30), 0.5 mg/ml (n=31), 0.5 mg/0.5 ml (n=23) and 0.25 mg/ml (n=30) GnRH antagonist (antide). Serum samples were taken three times daily during antide administration to assess antide and hormone levels. The minimal effective dose was defined as the lowest dose group with <2 LH surges (LH >12.4 IU/l and progesterone >2 ng/ml). RESULTS: Serum antide levels, mean LH and E2 levels per day and their area under the curves were dose-related to antide. The bioavailability of antide almost doubled after dilution in larger volumes. Pre-injection LH levels gradually increased during GnRH antagonist treatment. LH surges occurred in the lowest dose groups 0.5 mg/ml (3.2%), 0.5 mg/0.5 ml (6.7%) and 0.25 mg/ml (13.3%). Hence, 0.5 mg/ml is considered to be the minimal effective dose. Antide was overall well tolerated and safe. CONCLUSIONS: 0.5 mg/ml antide is the minimal effective dose to prevent an untimely LH surge in IVF patients stimulated with r-hFSH.

Key words: antide/dose-finding/GnRH antagonist/IVF/LH


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
GnRH agonists have been used in IVF cycles for many years to prevent premature LH surges. Placebo-controlled studies of GnRH agonists reveal that LH surges occur in ~20% of IVF patients, leading to cancellation of the IVF cycles (Edwards et al., 1996Go; Janssens et al., 2000Go). The long protocol became the standard in most centers and proved to be beneficial for the quality and number of retrieved oocytes, resulting in more embryos (Daya, 2000Go). In general, more embryos allow better selection so that the outcome, in terms of pregnancy rates, is improved (Templeton and Morris, 1998Go). It takes time to induce a state of desensitization after GnRH agonist administration. Unlike the agonists, GnRH antagonists cause an immediate and rapid, reversible suppression of gonadotropin secretion (Huirne and Lambalk, 2001Go). These compounds can be administered when premature luteinization during IVF stimulation is imminent, reducing the number of GnRH antagonist injections required. Antagonist therapy has proved to be a very convenient alternative to the long agonist strategy. The efficacy of this strategy, using single or multiple dosing regimen, has been demonstrated in several studies with ganirelix and cetrorelix (Ganirelix dose-finding Study Group, 1998Go; Albano et al., 1997Go; Olivennes et al., 1998Go). Both compounds belong to the so called third generation GnRH antagonists which exhibit low histamine releasing properties and can therefore be safely administered with low risks for allergic reactions (Bajusz et al., 1988Go; Rivier et al., 1992Go). Another compound belonging to this generation of GnRH antagonists is antide (Iturelix, Serono, Geneva, Switzerland) (Ljungqvist et al., 1988Go). Low histamine releasing properties in comparison to Ganirelix or Cetrorelix were reported, even lower than native GnRH (Bajusz et al., 1988Go; Ljungqvist et al., 1988Go; Nestor et al., 1992Go). The current study was designed to select the minimal effective dose of antide for the prevention of premature LH surges during ovarian stimulation when administered daily by subcutaneous (s.c.) injections. Efficacy was assessed in terms of hormone suppression, IVF outcome and safety aspects. This report will focus on the hormone and antide levels in the various dose groups. The effect of various induced LH levels on IVF outcome will be reported elsewhere (Huirne et al., submitted).


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Patients
All patients scheduled for IVF or ICSI treatment in our department during a period of 7 months were assessed for eligibility. A total of 144 patients participated in the study. All patients were aged between 21 and 39 years, had spontaneous regular menstrual cycles between 25 and 35 days, two ovaries, a normal uterine cavity, a body mass index ≤30 and had at least two spontaneous menstruations after the last clomiphene citrate or gonadotropin treatment. Women with elevated hormone levels (FSH ≥10 IU/l or LH ≥8 IU/l or prolactin levels (≥800 mIU/l) on cycle day 2 or 3 were excluded from participation. Patients with polycystic ovaries syndrome (defined as oligomenorrhoe and elevated LH levels or signs of hyperandrogenism) were also excluded. Also excluded were patients with abnormal hematological or biochemical parameters, patients with any previous ART cycle with fewer than 3 oocytes, known allergy or hypersensitivity to human gonadotropin preparations or GnRH analogues. The protocol was approved by the Committee on Ethics of Research involving Human Subjects of the Vrije Universiteit medical center (VUmc), Amsterdam, The Netherlands. All participants signed informed consent.

Study design
A phase II, single center study, conducted in two phases: a double-blind phase with two parallel treatment groups was followed by an open phase (see Figure 1). In the double-blind phase 60 patients were randomized to two different treatment groups (A: 2 mg/2 ml, B: 1 mg/ml). To improve patients' convenience, the 2 mg in group A was given as two injections of 1 mg/ml antide, one injection in the morning and one in the evening, since we expected that one injection with a volume of 2 ml would be too painful. Patients in group B received placebo in the morning and 1 mg/ml antide in the evening. Since none of the two groups turned out to be a failure group (i.e.with ≥2 LH surges) we decided to add an open phase in which three additional treatment groups with lower antide dosages were studied (0.5 mg/0.5 ml, 0.5 mg/ml, 0.25 mg/ml), administered once daily. The additional arms were added in a consecutive order and patients were enrolled in a chronological way. New evidence which became available after the start of this study suggested that the bio-availability of antide increases after dilution in larger volumes of glucose 5% (Data on file, Serono, Geneva). Therefore in two groups (0.5 mg/ml and 0.25 mg/ml) antide was diluted in larger volumes of glucose 5% solution; 0.5 mg and 0.25 mg in 1 ml, respectively. This means that in two arms 0.5 mg antide was administered but in group C it was diluted in 1.0 ml of glucose 5% solution and in group D it was diluted in 0.5 ml of glucose 5% solution.



View larger version (42K):
[in this window]
[in a new window]
 
Figure 1. Flow diagram, showing the flow of the participants through each stage of the trial. RFSH, recombinant FSH; antide, GnRH antagonist; rHCG, recombinant HCG; OPU, ovum pick-up; ET, embryo transfer. All patients included had follow-up of serum samples throughout the entire treatment period (three times daily) and during ET, no serum values were missing.

 
Each group was intended to contain 30 patients unless more than 1 LH surge occurred, which according to the protocol lead to the discontinuation of that particular dose group and was considered to be a failure dose. More than 1 LH surge per 30 patients was considered to be unacceptable for clinical use in IVF patients.

Masking
Treatment packs for the double-blind phase of the study were prepared according to the randomization list by Serono International (Geneva, Switzerland). Patient packs, containing antide/placebo or antide/antide vials, were labeled with unique study identification numbers, provided by Serono International (Geneva, Switzerland); placebo vials contained a sterile isotonic aqueous solution. When eligible, patients were enrolled to the study by one of the two responsible trained researchers and received a unique study number in a chronological order at the start of the first stimulation day. The code was not known to the executors of the study. Assignment to group A or B was therefore double blind, assignment to group C, D or E was dependent on the chronological entry of the study.

Treatment protocol
The treatment protocol is illustrated in Figure 2. On day 2 or 3 of a spontaneous menstruation, r-hFSH (Gonal-F®, Serono, Aubonne, Switzerland) was given as a single daily s.c. injection. The starting dose varied between 150–300 IU, depending on previous ovarian response, but was fixed for the first 5 days. After this period, depending on ovarian response as assessed by daily ultrasound, the r-hFSH dose could be adjusted. All antide, placebo and r-hFSH injections from stimulation day 6 (S6) onwards, were given subcutaneously, by a trained research professional. From stimulation day 6 onward, up to and including the day of recombinant human chronic gonadotropin (r-hCG; Ovitrelle®, Serono) administration, daily antide was given. Recombinant hCG was administered as soon as one follicle ≥18 and two follicles were ≥16 mm. Thirty-six hours after r-hCG administration, ovum pick-up (OPU) was performed transvaginally and ultrasound guided. The OPU was followed by IVF with or without ICSI, a maximum of three embryos were replaced 2–3 days thereafter. Luteal support (200 mg progesterone vaginally, three times daily) was started 1 day after OPU until the third week of pregnancy or a negative pregnancy test.



View larger version (13K):
[in this window]
[in a new window]
 
Figure 2. Schematic overview of the treatment schedule. Schematic overview of the treatment regimen with a GnRH antagonist (antide) in patients undergoing ovarian hyperstimulation with r-hFSH (Gonal-f®). S1, stimulation day 1; S6, stimulation day 6; OPU, ovum pick-up; ET, embryo transfer.

 
Assessments
One to three months before randomization, serum samples were taken to assess haematology (haemoglobin, haematocrite, white cell count, red cell count and platelets), biochemistry (sodium, potassium, creatinine, total bilirubin, calcium, total cholesterol, alkaline phosphatase, SGPT, LDH, total protein) and hormone levels (FSH, LH, oestradiol, progesterone and prolactin), taken on cycle day 2 or 3. Urine was checked for glucose, ketones, haematuria and proteinuria. On S1 (stimulation day 1), before any study drug was administered, a blood sample was taken to perform a pregnancy test and to assess FSH, LH, oestradiol (E2) and progesterone (P4) levels and a transvaginal ultrasound was performed to measure follicular activity, endometrial thickness and to exclude the presence of cysts. During antide administration, three samples per day were taken (in the morning before any injection, in the evening prior to Gonal-f or antide injection, and 20–84 min later) to assess serum levels (FSH, LH, E2, P4 and antide). The potential variation in timing of the evening post-injection bloodsampling was intended to allow pharmacokinetic and pharmcodynamic modeling. The mean sample time was 34.6 (SD 3.9) min after antide injection (range of the mean per patient varied from 30 to 53 min). The transvaginal ultrasound was performed daily to assess follicular development and endometrial thickness. On the day of embryo transfer (ET), serum samples were taken to measure the antide level, and 7–11 days after OPU to measure the levels of progesterone and antide. The prestudy haematological and biochemical parameters were reassessed after the treatment cycle to evaluate potential changes. Finally, 23–25 days after ovum pick-up, serum hCG levels were measured. If positive, a vaginal ultrasound was performed 35 and 42 days following rhCG administration, to record the number of fetal sacs and fetal heart activity. Ultrasound was repeated at a gestational age of 12 weeks. The pregnancy results will be reported elsewhere (Huirne et al., submitted).

Local tolerance after the different injections was assessed 60 min and 12 h after each antide injection. Pain, itching, tenderness, redness, swelling and bruising were recorded on a 4 point scale (none, mild, moderate and severe) of a diary card. All other side effects were reported daily on this diary card as well.

Serum assessment
Blood samples were processed to serum immediately after collection and stored at –20°C. Routine haematology, biochemistry and urine assessment were performed by the local laboratory (The Central Laboratory of the VUmc) using commercially available immunometric assays. To detect LH surges, all morning samples were assessed daily by this local laboratory to measure the LH levels, using immunometric assay kits from Amerlite (Amerlite, Amersham, Bucks, UK). An LH surge was defined as LH >10 IU/l and P4 >2 ng/ml using this assay. Halfway through the study we were forced to change the assay, since Amerlite assays were not available anymore. We decided to use Delphia (Dephia, Wallac, Finland) assays. During the transition period of the assays, we assessed LH levels using both assays in 89 patients. Excellent correlation was observed between the two assays (R=0.98). A regression analyses revealed that the coefficient of LH was 1.24 using Delphia assay in comparison to Amerlite assay, thus the LH cut-off level 10 IU/l assessed by Amerlite was equivalent to 12.4 IU/l if assessed by Delphia assay.

For definitive analyses of all hormone and antide levels, as presented in this report, all serum samples (taken three times daily) were assessed retrospectively by LCG Bioscience Services LTD. E2 was measured using Sorin Radioimmunoassay, P4 using DPC Coat-a-count, RIA solid phase coated tube separation, FSH and LH using Serono MAIAclone IRMA. The lower limit of quantification for LH was 1 IU/l. For the retrospective analyses we defined an LH surge as LH >12.4 IU/l and P4 >2 ng/ml in one or more samples, taking all samples (three times daily) into account from S6 until hCG administration day, equivalent to the cut-off levels using the Delphia assays. The retrospective centralized analysis of serum antide levels were performed by Woods Assay (RIA), all samples were analyzed in triplicate, 1 µg/l is the Limit of Quantification.

Outcome measures
The primary endpoint of this study was to determine the minimal effective dose, defined as the lowest dose group in which fewer than two LH surges occurred. Secondary endpoints were drug requirements, serum hormone and antide levels and safety aspects.

Statistical analysis
The treatment groups were compared depending on the nature of the variables, [i.e the analysis of variance (ANOVA) or the analysis of covariance (ANCOVA), Chi-square tests, Fisher's exact test or non-parametric ranking methods like Kruskall–Wallis and Mann–Whitney U tests]. Results are reported as mean±SD. Correlations were calculated using Pearson's correlation coefficient. P<0.05 was considered to be statistically significant. Analyses were conducted on an intention to treat base. An overall dose-response test for trend with the treatment groups were performed on all efficacy data. The statistical hypothesis of no treatment effect was tested against the ordered alternative hypothesis that at least one antide dose is superior to the lowest dose group and that the response is decreased, or at least equal to those of the previous group when decreasing the dose. The statistical methods used to test this hypothesis were directly related to the nature of the variable, i.e. ANOVA with linear contrast, Jonckheere–Terpstra test or a Cochran–Armitage test. To identify different covariates on pharmacokinetic parameters, univariate analysis within NONMEM was used.

Total exposure to antide and hormone levels was expressed as area under the curves (AUC) during antagonist administration (i.e. S6-hCG day). For this calculation, the sum of the mean daily levels of all days during antide treatment were taken.

To calculate the induced change in serum levels in comparison to the basal level on stimulation day 6 (the moment on which the antagonist was started), AUC was calculated after subtraction of the basal level on S6 of all samples, defined as AUC–S6.

The sample size was not calculated; we based our groups size on clinically relevant arguments. More than one LH surge per 30 patients was considered to be unacceptable for clinical use in IVF patients. Therefore the number of patients was intended to be 30 per group, unless more than one LH surge occurred.


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Patients' characteristics
Baseline characteristics were similar between the five groups, except for type of fertility disorder (see Table I). Baseline biochemistry and hematology parameters in serum and urine were comparable between the five groups (data not shown), as were baseline FSH, LH and progesterone levels. The baseline oestradiol levels were significantly higher in group C (0.5 mg/ml) and E (0.25 mg/ml) in comparison to groups A (2 mg/2 ml) and B (1 mg/ml).


View this table:
[in this window]
[in a new window]
 
Table I. Baseline characteristics

 
Number of patients per treatment stage
Sixty patients (30 in each group) were enrolled in the double blind phase. In the open phase, 31 patients were successfully enrolled in group C (0.5 mg/ml). Group D failed after the inclusion of 23 patients, since two LH surges occurred. In the last group (E: 0.25 mg/ml) a total of four LH surges occurred. At the time of the second LH surge, which according to the protocol leads to the discontinuation of the treatment arm, the last five patients of this treatment group had already started with antide and were allowed to finish the started treatment. Two additional LH surges occurred in the last five patients of this arm. All patients received antide and r-FSH. One patient (group D: 0.5 mg/0.5 ml) did not receive r-hCG and had no ovum pickup, because of premature ovulation. Six additional patients did not undergo an embryo transfer because of total fertilization failure (one in 1 mg/ml, one in 0.5 mg/ml, two in 0.5/0.5 mg and two in 0.25 mg/ml group).

Stimulation phase
The mean duration of antide and r-hFSH administration was 5 (range 2–9) and 9 days (range 6–13), respectively. This was comparable between the groups (P=0.89) (see Table II). The mean Gonal-F® starting dose was similar in the various dose groups and varied between 223 and 238 IU (P=0.78). Additionally, the mean dose Gonal-F® per day and the mean total dose of Gonal-F® were also similar in the various dose groups (P=0.98 and 0.91, respectively). Because of insufficient response, the Gonal-F® dose was more often increased in the higher antide dose groups (P=0.04).


View this table:
[in this window]
[in a new window]
 
Table II. Drug requirements

 
Serum antide levels
The daily antide levels, the mean antide levels per day and the total AUC of antide during treatment were different between the various treatment groups (P<0.001) (see Table III and Figure 3). A two-compartment disposition model with first order absorption and first order elimination best described the pharmacokinetics of antide. The residual error model was a combination of an additive and a proportional part. The absorption half-life was ~10 min. The calculated bioavailability was different between the treatment groups: the higher the concentration the lower the relative bioavailability data (data not shown). The dilution of 0.5 mg antide in 1.0 ml (group C) versus 0.5 ml (group D) increased the mean antide levels/day from 0.56 to 0.87 µg/l, respectively, (P=0.01) (see Figure 3). The antide serum levels were below the limit of detection on the day of embryo transfer and during the post treatment visit. Using univariate analysis within NONMEM, body weight was found to linearly affect the apparent clearance, the apparent central volume and the absorption rate constant (data not shown). BMI also affected the pharmacokinetic parameters of antide but at a lower degree of significance compared to body weight. After the full covariate selection was performed, only the effect of body weight on apparent clearance and the apparent central volume remained in the final, model. Mean weight was 66.2 kg (range 45–94 kg), and mean height 167.7 cm (range 150–186), no differences were found between the various dose groups.


View this table:
[in this window]
[in a new window]
 
Table III. Serum antide levels

 


View larger version (14K):
[in this window]
[in a new window]
 
Figure 3. Mean antide serum levels per day. Box and wisker plot of the mean antide levels per day in the various treatment groups; A (2 mg/2 ml, B (1 mg/ml), C (0.5 mg/ml), D (0.5 mg/0.5 ml), E (0.25 mg/ml). The median levels were different between all treatment groups, P<0.001 (Mann–Whitney U). The trend for linear association with the dose is highly significant P<0.001 (Jonckheere–Terpstra). Dots above the box plots are the outliers.

 
LH levels during stimulation
LH levels immediately decreased after the first antide injection. LH values in the various treatment groups are presented in Table IV. The LH levels on S8 and the LH-AUC were different between the antide dose groups (P=0.03 and <0.001, respectively; Kruskall–Wallis) and were dose related (both P<0.001; Jonckheere–Terpstra). Thus high LH levels were associated with low antide levels. The differences in LH levels induced by antide, expressed by LH-AUC–S6 were also dose-related to the treatment groups (P<0.001).


View this table:
[in this window]
[in a new window]
 
Table IV. Gonadotropin serum levels

 
A gradual increase of the mean pre-injection LH levels were observed from S7 to the day of r-hCG (Figure 4). Overall, the mean LH serum levels increased with 0.76 IU/l (95%CI: 0.45–1.07). This increase was clearest in the lowest dose group (P=0.04), but was also present in some individual plots in the higher dose groups (see Figure 5).



View larger version (30K):
[in this window]
[in a new window]
 
Figure 4. Mean serum levels: FSH, LH, E2 and P4. Mean serum LH, FSH, E2 and P4 levels: pretreatment levels (on cycle day 2 or 3 before the study = PRE), on stimulation day 1 (CD2 or 3 during the study, before any medication was given) and daily from stimulation day 6 (S6) up to and including stimulation day 9 (S9) in the morning (am) and evening (pm).

 


View larger version (18K):
[in this window]
[in a new window]
 
Figure 5. Plot of LH and antide levels in an individual patient. Individual LH and antide levels from S6 to rhCG day of a patient receiving 1 mg/ml antide.

 
LH surges
Before antide administration (on S6) one patient (in the 0.5 mg/ml group) had a LH of 15.1 IU/l with a progesterone of 1 ng/ml; after antide administration the LH levels decreased to 3.8 IU/l. During antide treatment, eight patients had a LH >12.4 IU/l and seven had a concomitant P4 > 2 ng/ml before the criteria of hCG administration were met; one (LH 13.1 IU/l and P4 2.0 ng/ml) in the 1 mg/ml group, two in the 0.5 mg/0.5 ml group (LH 15.2 IU/l and P4 2.6 ng/ml, LH 19.4 IU/l and P4 4.4 ng/ml), none in the 0.5 mg/ml group (one patient had a LH of 14.1 but a P4 of 1.8 ng/ml) and four in the 0.25 mg/ml group (LH 13.9 IU/l and P4 3.6 ng/ml, LH 21.0 IU/l and P4 3.9 ng/ml, LH 32.0 and P4 2.0 ng/ml, LH 12.3 IU/l and P4 30 ng/ml). The incidence of LH surges was significantly different between the dose groups (P=0.04).

FSH, oestradiol and progesterone levels during stimulation
On the day of hCG, the FSH levels were higher in group A (2.0 mg/ml) and group B (1.0 mg/ml) in comparison to the lower dose groups (P=0.05). The total AUC of the FSH levels were not significantly different between the individual treatment groups (P=0.25) (see Table IV).

The steroid hormone levels are presented in Table V. E2 levels increased in all groups during antide administration (see Figure 4). The mean E2 levels/day and the induced E2 AUC, adjusted for baseline levels on S6 (E2-AUC–S6) were proportionally dose-related to the treatment group (P=0.009 and <0.001, respectively), with the highest E2 levels in the lowest antide dose groups. The same was so for the calculated E2 levels per follicles >10 mm on the day of hCG administration, with the highest E2 levels in the lowest dose group (0.25 mg) and the lowest E2 levels in the highest antide dose group (2.0 mg), 501 (228) and 388 (168) pmol/l/follicle, respectively (P=0.05).


View this table:
[in this window]
[in a new window]
 
Table V. Serum steroid hormone levels

 
On all days, the mean P4 levels were comparable in the various dose groups. Progesterone levels showed large fluctuations during the day (see Figure 4). All calculated AUC or induced differences in P4 levels (P4-AUC, P4-AUC–S6) did not differ between the various treatment groups and the P4 levels were not significantly correlated to the actual antide levels. But linear regression analyses of the P4 AUC (P<0.001, R2=0.36) revealed that, if adjusted for the duration of the treatment, the absolute P4 levels were dependent on LH-AUC ({beta}1=0.61, P<0.001) and the number of follicles at hCG day ({beta}2=0.20, P=0.004).

Safety and tolerance
Overall, antide was well tolerated, with none of the patients experiencing a serious adverse event related to medication. The most frequently recorded side effects were general disorders (especially fatigue and headache) and gastro-intestinal disorders (abdominal pain and nausea). 83.1% of all patients recorded at least one local skin reaction (redness, tenderness, itching, bruising or swelling) with transient redness (25%) and tenderness at the injection site (22.2%) being most frequent. Most patients considered the local reactions to be mild. The local reactions mostly resolved spontaneously within 1 h.8.8% of patients had at least one moderate local reaction occurring 1 h after antide injection and were mostly found in treatment groups A and B. No severe local reactions were observed 1 h after antide injection. In the minimal effective dosegroup (D: 0.5 mg/ml), a moderate local reaction observed 1 h after the injections was observed in only one patient. The biochemical and hematological parameters measured in all patients were similar before and after the treatment cycle (data not shown).


    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
In this study we established that a daily subcutaneous dose of 0.5 mg/ml antide, when given from the 6th day of stimulation with r-hFSH, is sufficient to prevent significant rises in LH and progesterone in IVF/ICSI. We confirm that the approach of a daily small dose of a GnRH antagonist is useful in IVF/ICSI stimulation protocols to avoid cancellation of the procedure due to premature lutinization that otherwise would occur in ~20% of all started cycles (Janssens et al., 2000Go).

We used the same standard definition for premature luteinization as others in GnRH antagonist dose-finding studies, namely a rise of LH above a predetermined threshold plus a significant increase in serum progesterone values (Albano et al., 1997Go; Ganirelix dose-finding Study Group, 1998Go). According to our definitions, groups D (0.5 mg/0.5 ml) and E (0.25 mg/ml) were considered to be failure dosages, and group C (0.5 mg/ml) appeared to be the minimal effective dose, in which no rises occurred. It should be noted that sampling blood three times per day increases the detection of any significant changes in LH. The design of the study was chosen on ethical grounds, with the aim of minimal exposure to possible premature luteinization. We were aware of the risk of a possible imbalance in baseline characteristics. Retrospectively it appeared this was present as regards type of fertility disorder and pretreatment oestradiol levels between the various groups. Nevertheless, these factors had no effect on pharmacokinetics and hormonal dynamics.

As expected there was a clear dose dependency of LH secretion throughout the stimulation period with lowest levels with the highest antide dosages. Similar dose dependency was seen in the other GnRH antagonist dose-finding studies (Albano et al., 1997Go; Ganirelix dose-finding Study Group, 1998Go). This study clearly demonstrated that the dilution of 0.5 mg antide in a larger volume of glucose 5% increased the bioavailability of antide and resulted in a lower incidence of LH surges. The antide serum levels were almost doubled if the same dose of antide (0.5 mg) was diluted in 1 ml instead of 0.5 ml.

A clear and intriguing observation in our study was the remarkable rise of preinjection LH levels throughout the antide treatment period. It occurred with all dosages, but was particularly clear in the lowest antide dose groups. This gradual increase of basal LH secretion while daily injections of a GnRH antagonist are given seems a common feature with this treatment strategy (Ganirelix dose-finding Study Group, 1998Go). It can also be seen in healthy unstimulated women (Oberye et al., 1999Go). Apparently some escape of pituitary LH secretion takes place, which even occurs while serum concentrations of antide remain constant. One explanation is that the pituitary develops an increased sensitivity to endogenous hypothalamic GnRH under these circumstances. The mechanism by which this takes place remains obscure. Possibly changes in hormonal milieu play a role. Increasing oestradiol levels during stimulation may modify pituitary sensitivity. The gradual escape was indeed most prominent in the lowest antide dose groups, which was associated with the highest oestradiol levels, both in our study and also in the Ganirelix dose-finding study (Ganirelix dose-finding Study Group, 1998Go). Alternatively, pituitary GnRH receptors may have been up-regulated by the GnRH antagonist itself as suggested by several authors (Gordon et al., 1994Go). A final explanation for the LH increase may be that the GnRH antagonists directly intervene in the ultrashort-loop feedback mechanisms comprising GnRH autoregulation (Krsmanovic et al., 1999Go).

The FSH levels on the day of hCG were higher in the highest antide dose groups, while there was a trend of a dose dependency in mean daily FSH levels. The adjustments of daily FSH dose, based on insufficient follicular response, may have been responsible for this. Although similar total FSH quantities were used in all antide dose groups, significantly more patients had r-hFSH dose increases in the highest antide dose groups.

The mean oestradiol levels per day and the induced changes in oestradiol levels during antide administration in comparison to the baseline levels, were inversely related to the antide dose. This is in agreement with an earlier report (Ganirelix dose-finding Study Group, 1998Go). Similarly, the highest levels of oestradiol production per follicle ≥11 mm were found in the lowest antide dose group. The dose-related oestradiol response to the GnRH antagonist, independent from the number of follicles, indicates a decrease of its production by granulosa cells, probably as a result of reduced availability of androstendione substrate.

It is often suggested that late follicular progesterone secretion is LH dependent. However, a straightforward LH dependency of progesterone secretion was not obvious in our study. Nevertheless, this relation became clear after correction for numbers of follicle on the day of hCG. The total AUC of progesterone increased by 0.6 ng/ml when LH-AUC increased by 1 IU/L. Apparently, the absolute level of progesterone production is also dependent on the total number of follicles. This means that high progesterone levels, independent of high LH levels, may also be the result of profound follicular growth in the ovary.

Overall, antide was well tolerated and safe in terms of side effects and influence on biochemical and haematological parameters. The reported reactions were all minor and most resolved spontaneously within 1 h. Only 9% of the patients had a moderate or severe local skin reaction 1 h after antide administration during the entire treatment period. This was lower than with other antagonist studies employing the same assessment methods, scales and criteria for these side effects (Ganirelix dose-finding Study Group, 1998Go; Middle East Orgalutran Study Group, 2001Go). This can be explained by the much lower histamine releasing property of antide compared to the clinically available GnRH antagonists (Bajusz et al., 1988Go; Rivier et al., 1992Go).

In conclusion, 0.5 mg/ml antide is the minimal effective dose to prevent LH surges in hyperstimulated cycles for IVF or ICSI. LH levels increase gradually during GnRH antagonist treatment. The bioavailability increases if antide is diluted in a larger volume. Antide was well tolerated and safe.


    Acknowledgements
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
The authors would like to thank Ted Korsen and Agnes Wierink-Wanders, research nurses for their very active support to this study; Carien van de Brand and Floris holzheimer, Paxel Mirai, for the monitoring of this study. This study was sponsored by Serono International, Geneva, Switzerland.


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Albano C, Smitz J, Camus M, Riethmuller-Winzen H, Van Steirteghem A and Devroey P (1997) Comparison of different doses of gonadotropin-releasing hormone antagonist Cetrorelix during controlled ovarian hyperstimulation. Fertil Steril 67, 917–922.[CrossRef][ISI][Medline]

Bajusz S, Kovacs M, Gazdag M, Bokser L, Karashima T, Csernus VJ, Janaky T, Guoth J and Schally AV (1988) Highly potent antagonists of luteinizing hormone-releasing hormone free of edematogenic effects. Proc Natl Acad Sci USA 85, 1637–1641.[Abstract]

Daya S (2000) Gonadotropin releasing hormone agonist protocols for pituitary desensitization in in vitro fertilization and gamete intrafallopian transfer cycles (Cochrane review). In: the Cochrane library, issue 1. Oxford Update Software.

Edwards RG, Lobo R and Bouchard P (1996) Time to revolutionize ovarian stimulation. Hum Reprod 11, 917–919.[ISI][Medline]

European and Middle East Orgulatran Study Group (2001) Comparable clinical outcome using the GnRH antagonist ganirelix or a long protocol of the GnRH agonist triptorelin for the prevention of premature LH surges in women undergoing ovarian stimulation. Hum Reprod 16, 644–651.[Abstract/Free Full Text]

Ganirelix dose-finding Study Group (1998) A double-blind, randomized, dose-finding study to assess the efficacy of the gonadotrophin-releasing hormone antagonist ganirelix (Org 37462) to prevent premature luteinizing hormone surges in women undergoing ovarian stimulation with recombinant follicle stimulating hormone (Puregon). Hum Reprod 13, 3023–3031.[Abstract]

Gordon K, Scott RT, Williams RF, Danforth DR, Loozen HJ, Kloosterboer HJ and Hodgen GD (1994) In vivo effects of a potent GnRH antagonist ORG 30850: physiologic evidence that down-regulation of GnRH receptors does not occur. J Soc Gynecol Investig 1, 290–296.[ISI][Medline]

Huirne JAF, Van Loenen ACD, Schats R, McDonnel J, Hompes PGA, Schoemaker J, Homburg R and Lambalk CB. Dose-finding study of daily gonadotropin-releasing hormone (GnRH) antagonist for the prevention of premature luteinizing hormone surges in IVF/ICSI: optimal changes in LH and progesterone. Hum Reprod (submitted).

Huirne JAF and Lambalk CB (2001) Gonadotropin-releasing-hormone-receptor antagonists. Lancet 358, 1793–1803.[CrossRef][ISI][Medline]

Janssens RM, Lambalk CB, Vermeiden JP, Schats R, Bernards JM, Rekers-Mombarg LT and Schoemaker J (2000) Dose-finding study of triptorelin acetate for prevention of a premature LH surge in IVF: a prospective, randomized, double-blind, placebo-controlled study. Hum Reprod 15, 2333–2340.[Abstract/Free Full Text]

Krsmanovic LZ, Martinez-Fuentes AJ, Arora KK, Mores N, Navarro CE, Chen HC, Stojilkovic SS and Catt KJ (1999) Autocrine regulation of gonadotropin-releasing hormone secretion in cultured hypothalamic neurons. Endocrinology 140, 1423–1431.[Abstract/Free Full Text]

Ljungqvist A, Feng DM, Hook W, Shen ZX, Bowers C and Folkers K (1988) Antide and related antagonists of luteinizing hormone release with long action and oral activity. Proc Natl Acad Sci USA 85, 8236–8240.[Abstract]

Nestor JJ, Jr, Tahilramani R, Ho TL, Goodpasture JC, Vickery BH and Ferrandon P (1992) Potent gonadotropin releasing hormone antagonists with low histamine-releasing activity. J Med Chem 35, 3942–3948.[ISI][Medline]

Oberye JJ, Mannaerts BM, Huisman JA and Timmer CJ (1999) Pharmacokinetic and pharmacodynamic characteristics of ganirelix (Antagon/Orgalutran). Part II. Dose-proportionality and gonadotropin suppression after multiple doses of ganirelix in healthy female volunteers. Fertil Steril 72, 1006–1012.[CrossRef][ISI][Medline]

Olivennes F, Alvarez S, Bouchard P, Fanchin R, Salat-Baroux J and Frydman R (1998) The use of a GnRH antagonist (Cetrorelix) in a single dose protocol in IVF-embryo transfer: a dose finding study of 3 versus 2 mg. Hum Reprod 13, 2411–2414.[Abstract]

Rivier J, Porter J, Hoeger C, Theobald P, Craig AG, Dykert J, Corrigan A, Perrin M, Hook WA and Siraganian RP (1992) Gonadotropin-releasing hormone antagonists with N omega-triazolylornithine, -lysine, or -p-aminophenylalanine residues at positions 5 and 6. J Med Chem 35, 4270–4278.[ISI][Medline]

Templeton A and Morris JK (1998) Reducing the risk of multiple births by transfer of two embryos after in vitro fertilization. N Engl J Med 339, 573–577.[Abstract/Free Full Text]

Submitted on February 5, 2004; accepted on May 14, 2004.