Hum. Reprod. Advance Access published online on May 15, 2008
Human Reproduction, doi:10.1093/humrep/den131
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Short-term androgen priming by use of aromatase inhibitor and hCG before controlled ovarian stimulation for IVF. A randomized controlled trial
1 The Fertility Clinic, Section 4071, Copenhagen University Hospital, Rigshospitalet, DK-2100 Copenhagen, Denmark 2 Laboratory of Reproductive Biology, Section 5712, Copenhagen University Hospital, Rigshospitalet, DK-2100 Copenhagen, Denmark
3 Correspondence address. E-mail: kristine.loessl{at}rh.dk
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and MethodsResultsDiscussionFundingAcknowledgementReference |
|---|
BACKGROUND: Temporary exposure of follicles to increased levels of androgens may augment follicular responsiveness. The present study tested whether short-term androgen priming by aromatase inhibitor and human chorionic gonadotrophin (hCG) before controlled ovarian stimulation (COS) increases the number of top-quality embryos after IVF/ICSI.
METHODS: Patients were randomized to androgen priming (n = 53): anastrozole 1 mg cycle day (c.d.) 2, 3 and 4, hCG 1250 IU and cetrorelix 3 mg on c.d. 2, rFSH 150 IU from c.d. 5 following a flexible antagonist protocol; or control (n = 50): flexible antagonist protocol.
RESULTS: The mean (confidence interval) number of top-quality embryos was 1.08 (0.83,1.40) and 1.43 (1.12,1.81) in the priming and control group, respectively, being 32% (–7%, 89%) higher in the control compared to priming group (P = 0.120). Stimulation duration was longer in the priming group (P < 0.001). On the day of hCG administration, the proportion of c.d. 2 antral follicles reaching 
14 mm was higher in the priming group (P = 0.014), as were serum estradiol (E2) (P < 0.001) and E2 per follicle 14 mm (P = 0.005). Pre-ovulatory follicular fluid levels of E2 (P = 0.007) and testosterone (P = 0.014) were higher in the priming group. The number of oocytes retrieved was similar. The fertilization rate was lower in the priming group (P = 0.007). Ongoing pregnancy rates in priming and control group were 30 and 36% (P = 0.531).
CONCLUSIONS: Administration of aromatase inhibitor and hCG before COS for IVF/ICSI failed to improve the number of top-quality embryos. ClinicalTrials.gov Identifier is NCT00286364 [ClinicalTrials.gov] .
Key words: androgens/aromatase inhibitor/follicular fluid/hCG/IVF
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsDiscussionFundingAcknowledgementReference |
|---|
Aromatase inhibitors prevent the conversion of androgens into estrogens by a specific inhibition of the aromatase enzyme (P450arom). These drugs are used primarily in the treatment of hormone sensitive breast cancers in post-menopausal women. Since 2001, aromatase inhibitors have been used on an experimental basis for ovulation induction and for controlled ovarian stimulation (COS) in infertile women due to the drugs' ability to increase endogenous gonadotrophin levels by blockage of estrogen synthesis (Mitwally and Casper, 2001
).
We have studied the use of aromatase inhibitors to temporarily increase the local, intraovarian androgen level before COS, in order to improve follicular responsiveness by increasing the sensitivity of granulosa cells toward follicle- stimulating hormone (FSH). Animal studies have indicated that androgens regulate granulosa cell function of small antral follicles by an up-regulation of the expression of FSH receptor. The level of messenger RNA (mRNA) encoding for the FSH receptor was increased by exposure of androgen of granulosa cells from antral follicles in both rhesus monkey (Weil et al., 1999) and cow (Luo and Wiltbank, 2006). Further, short-term androgen administration has been shown to increase the number of small antral follicles as well as proliferation of granulosa and theca cells in the rhesus monkey (Vendola et al., 1998).
In the clinical setting of women undergoing ovarian stimulation for in vitro fertilization (IVF) or intracytoplasmatic sperm injection (ICSI), an up-regulation of granulosa cell FSH receptor just before ovarian stimulation could enhance follicular responsiveness to exogenous gonadotrophin and thus allow a larger proportion of the antral follicles to escape atresia and develop into mature follicles. Further, an increased expression of follicular FSH receptor could lead to enhanced induction of luteinizing hormone (LH) receptors in mid-late follicular stage (Rani et al., 1981; Shaw et al., 1989), which could in turn lead to better final oocyte maturation in response to LH/human chorionic gonadotrophin (hCG) hormone administration. This may result in oocytes with an improved developmental potential and hence more top-quality embryos per treatment cycle.
In a preliminary study, we developed the concept of androgen priming (Lossl et al., 2006). Two measures were employed to achieve a temporary increase in the exposure of the follicles to androgen. An aromatase inhibitor was used to prevent the aromatization of androgens into estrogens, and a bolus of hCG was given to promote theca cell androgen synthesis. During the priming period a gonadotrophin-releasing hormone (GnRH) antagonist was co-administered to maintain low levels of gonadotrophins and avoid direct stimulation of the follicles before possible androgen priming effects had occurred. The results indicated that short-term use of aromatase inhibitor and hCG prior to COS increased the local, intraovarian levels of androgen and that the concept may be of potential clinical use. The present study assessed whether short-term androgen priming by aromatase inhibitor and hCG before COS increases the number of top-quality embryos after IVF/ICSI.
|
Material and Methods |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsDiscussionFundingAcknowledgementReference |
|---|
Study design
This prospective randomized controlled single-center study was performed between September 2005 and 2006 at The Fertility Clinic, Copenhagen University Hospital, Rigshospitalet, Denmark. The local Ethics Committee—J.nr. (KF) 02 271500 approved the study. Each subject gave written informed consent before participating in the study. The study was conducted according to ‘Good Clinical Practice’ (GCP) and monitored by an independent external monitor from the GCP-unit at the Copenhagen Region.
Subjects
The study flow chart is shown in Fig. 1. A total of 103 patients were included to obtain 100 patients ‘treated per protocol (TPP)’ who fulfilled the planned in- and exclusion criteria. Inclusion criteria: (i) indication for IVF or ICSI, (ii) age 20–39 years, (iii) regular menstrual cycle within the range of 21–35 days (intra-individual variation ±3 days), (iv) body mass index between 18 and 30 kg/m2, (v) a maximum of three earlier IVF/ICSI treatments and (iv) negative urinary pregnancy test on cycle day (c.d.) 2. Exclusion criteria: (i) polycystic ovary syndrome (PCOS) (The Rotterdam ESHRE/ASRM-Sponsered PCOS consensus workshop group, 2004), (ii) earlier ‘poor response’ with less than four follicles 15 mm after earlier COS, (iii) earlier participation in the trial, (iv) a history of cardiovascular, thrombo-embolic, hepatic or renal disease. In three cases, all from the control group, the following protocol deviations appeared: earlier poor response (n = 1), double dosage of rFSH during the first 5 days of ovarian stimulation (n = 1) and administration of hCG for final oocyte maturation 9 h too late (n = 1). The patient with earlier poor response again developed only two follicles and was converted to intrauterine insemination on request. She conceived and delivered a healthy child.
|
Treatment protocols
Fig. 2 illustrates the treatment protocols. Patients were randomized to a modified flexible antagonist protocol including a 3-day period of androgen priming during early-follicular-phase GnRH antagonist administration, just before initiating exogenous rFSH stimulation, or to a standard flexible antagonist protocol.
|
The ‘priming group’ received hCG (Pregnyl, Organon AS, Denmark) 1250 IU sc on c.d. 2, anastrozole (Arimidex, AstraZeneca AS, Denmark) 1 mg orally on c.d. 2, 3 and 4 and additionally cetrorelix (Cetrotide, Serono Europe Ltd, UK) 3 mg sc on c.d. 2 to inhibit follicular growth during the priming period. Ovarian stimulation was initiated on c.d. 5 according to the same flexible antagonist protocol as applied to the controls.
The ‘control group’ initiated ovarian stimulation on c.d. 2 (stimulation day 1) with a fixed daily dose of 150 IU rFSH (Gonal f, Serono Europe Ltd, UK). This dose could be slightly adjusted only from stimulation day 6 in steps of 75 IU every second day based on the physician's judgment on follicular development. Four physicians (investigators) performed all the scans and decisions regarding treatment. GnRH antagonist, Cetrotide 0.25 mg sc, was given daily when the leading follicle reached 14 mm. As soon as two follicles reached a mean diameter of 17 mm, 10 000 IU hCG (Pregnyl) was administered followed by oocyte retrieval 36 h later. Subsequently IVF or ICSI was performed, the latter if the total recovery of motile sperm cells was less than two million after a density gradient centrifugation in the present or a previous cycle. One or two embryos were transferred on Day 2. The decision to transfer one versus two embryos was determined by the guidelines by The Danish National Board of Health, where elective single embryo transfer should be done in patients below 37 years, with one or more top-quality embryos, undergoing their first or second IVF/ICSI cycle. Vaginal progesterone was administered daily for 2 weeks from the day of embryo transfer and was not prolonged in case of pregnancy. Serum beta hCG was measured 2 weeks after embryo transfer and a value >2 IU/l was considered as a positive test result. Detection of fetal heart activity by transvaginal ultrasound at 7–8 weeks was considered as an ongoing pregnancy.
Fertilization rate and embryo quality
Experienced embryologists blinded to treatment evaluated oocyte fertilization, subsequent cleavage stage and embryo morphology (degree of fragmentation, blastomere uniformity and multinucleation). According to the clinic's routine procedures, embryos were scored on agreement between two embryologists who simultaneously judged each embryo using x400 magnification and a connected monitor. Two-pronuclear oocytes as well as oocytes that subsequently cleaved were considered fertilized. The individual fertilization rate was calculated as fertilized oocytes divided by injected oocytes when ICSI was performed and by all intact oocytes retrieved when IVF was performed. Embryos with 4–6 equally sized blastomeres on Day 2 with 
20% fragmentation and no multinucleation were considered as top-quality embryos. Embryos with 2–6 equally or unequally sized blastomeres with 20% fragmentation and no multinucleation were considered as good-quality embryos. Embryos with >20% fragmentation and embryos containing multinucleated blastomere(s) were not transferred.
Ultrasonography
Transvaginal ultrasonography was performed on c.d. 2, on stimulation day 1 and 6 and thereafter every second day until hCG administration. The following were measured: (i) ovarian volume on c.d. 2, (ii) endometrial thickness (every examination) and (iii) number of antral follicles in the following size categories: 2–4, 5–7, 8–10 mm and follicle size >10 mm (every examination). Follicular size was determined as mean diameter. If the patient had a follicle larger than 10 mm on stimulation day 1 (two patients in the priming group and one patient in the control group) or a short menstrual cycle between 21 and 24 days (one patient from each group), extra ultrasonography was performed between stimulation day 1 and 6 to optimize initiation of GnRH antagonist. Ultrasonography was performed using a GE LOCIQ 5 Expert scanner using a transvaginal E8C probe, B-mode, 8.0 MHz.
Serum endocrinology
Serum concentrations of FSH, LH, progesterone, testosterone and estradiol (E2) were measured on c.d. 2 and stimulation day 1, 3, 6 and thereafter every second day until hCG administration. c.d. 2 corresponded to stimulation day 1 in the control group. Serum analyses were performed on fresh samples by electrochemiluminescence immunoassay (Roche Diagnostics, Mannheim, Germany) in Modular analytics E170 immunoassay analyser. Inter- and intra-assay coefficients of variation were 4.0 and 1.5% for FSH, 2.1 and 0.8% for LH, 4.8 and 2.9% for progesterone, 6.0 and 1.5% for testosterone and 4.9 and 1.8% for E2.
Follicular fluid samples
Two follicular fluid (FF) samples from each patient undergoing oocyte retrieval were collected. They included the first aspirated follicle >17 mm from each ovary that contained an oocyte and was not visibly contaminated with blood. An exception was made in six and five women from the priming and control group, respectively, where only one FF sample was collected, as no oocytes were collected from one of their ovaries due to (i) difficult anatomical location, (ii) interposition of intestine or large vessels or (iii) no development of mature follicles. Further, collection of FF was missed in one patient from the priming group and omitted in the two patients from the control group with protocol deviations undergoing oocyte aspiration (Subject section). The separate FF samples were centrifuged (1000 x g) to eliminate granulosa and blood cells and stored at –20°C until analyses. No samples were pooled.
Progesterone, testosterone, androstendione and E2 were measured using commercially available RIA kits (DSL-3400, DSL-43100, DSL-4000, DSL-3800, respectively, Diagnostic System Laboratories, Houston, TX, USA). Samples were diluted 1:961 (progesterone and E2), undiluted (testosterone) and 1:31 (androstendione) in steroid-free serum just prior to measurement. Anti-müllerian hormone (AMH) was measured in undiluted samples by the use of a specific ELISA kit according to the manufacturers' instructions (DSL-10-14400; Diagnostic System Laboratories). Inhibin-B was measured using a specific ELISA-kit according manufactures instruction (The Oxford Bio-innovation kit; Biotech-IgG, Copenhagen, Denmark). Before measurements the FF samples were diluted 1:520, 1:135 or 1:102 in a pool of serum obtained from postmenopausal women (who had no inhibin B activity).
Study endpoints
The primary endpoint was number of top-quality embryos after IVF/ICSI defined as embryos that on Day 2 after oocyte aspiration consisted of 4–6 equally sized blastomeres with 20% fragmentation and no multinucleation. The number of top-quality embryos was chosen as the primary endpoint, as this could reflect both qualitative and quantitative effects of enhanced follicular FSH sensitivity.
Secondary endpoints were (i) the number of antral follicles before and after the androgen-priming period, and their development in response to stimulation, (ii) the levels of FSH, LH, E2, testosterone and progesterone in serum during treatment, (iii) the levels of steroids, inhibin B and AMH in FF and (iv) standard IVF parameters such as stimulation duration, total dose of FSH, endometrial development, number of oocytes retrieved, fertilization, implantation and pregnancy rate.
Sample size calculation
Since sample size calculation gives only a rough indication of the necessary sample size, we chose for simplicity to treat a discrete numerical variable (number of top-quality embryos) as a continuous outcome measure. In our preliminary study, mean (SD) number of top-quality embryos per oocyte retrieval was 1.5 (0.88) in the priming group (aromatase inhibitor and hCG), and 0.7 (0.85) in the control group (Andersen and Lossl, 2007). A difference of 0.5 in the number of top-quality embryos was considered as highly clinical relevant, since only single and double embryo transfers were offered to women <40 years. A sample size of 100 patients, considering a two-sided significance level of 0.05 and 80% power, would allow us to detect the above mentioned difference.
Randomization
A randomization list generated by computerized block randomization with blocks of unequal sizes was provided by Department of Biostatistics, Institute of Public Health, University of Copenhagen. Numbered and sealed opaque envelopes were prepared from this list by a secretary, who was the only person with access to the list. The block sizes were not disclosed during the study. After oral and written informed consent these opaque envelops were opened in consecutive order by one of the investigators just before inclusion and the start of treatment on c.d. 2. Since only patients randomized to the priming group received 3 days of androgen priming, it was found unrealistic to truly blind patients and investigators. However, the embryologists that evaluated the embryos, and thereby the primary endpoint, were blinded to treatment. According to the protocol, the inclusion was stopped when 100 patients were TPP, i.e. after a total inclusion of 103 patients, no matter where in the block. This was resulted in a slightly unequal number of patients in the two groups.
Statistical analyses
Intention-to-treat (ITT, n = 103) analyses are presented. A two-sided P < 0.05 was considered statistically significant. TPP analyses (n = 100) gave similar results with no change of any P-value around the 0.05 level. No adjustments for multiplicity were made regarding secondary endpoints. Numerical data are presented as median or mean according to data distribution with corresponding 5th and 95th percentiles for description and 95% confidence interval (CI) for estimation. Group comparisons were made by the Mann–Whitney U-test or the unpaired T-test. If the distribution of data in one or both groups was not compatible with a normal distribution, when tested by the Shapiro–Wilk test and visual inspection of histograms, the nonparametric approach was chosen. Additionally, the number of top-quality embryos was treated as a Poisson distributed count, as the range was relatively small (0–6) with highest frequencies close to zero, and the mean values were compared for the two groups.
Regarding FF analyses, the hormonal level of each patient was determined as the mean level of the two FF samples obtained from each woman (n = 85). When only one FF sample was obtained (n = 11), the hormonal levels of this single sample represented the patient level. This data reduction was done to enable the use of the Mann–Whitney U-test or unpaired T-test to compare the two groups as mentioned above, and the nonparametric Spearman correlation coefficients (Spearman's rho) to investigate possible associations between the hormonal levels, without overestimating the number of independent observations. The between-group comparisons of FF endocrine levels were also performed using all the individual FF samples (n = 181, logarithmically transformed whenever the distributional assumptions were improved) in a Variance Component Model taking into account the pair wise correlation of FF data. The P-values obtained were similar for hormones and hormone ratios reported in Table IV (data not shown), which confirmed the findings of the simpler statistics.
|
In the priming group, more couples had male factor infertility and more, although not significant, underwent ICSI as fertilization method. To adjust for the possible influence of fertilization method on fertilization rate, a two-way ANOVA was performed including both treatment group and fertilization method as fixed factors in analysis of fertilization rate.
Categorical data were analysed by the Chi-squared test. Within-group analyses were performed by the use of the Wilcoxon Signed Rank test or paired T-test, depending on the compatibility of the distribution of paired differences with a normal distribution.
The software packages SPSS, version 14 (SPSS Inc., Chicago, USA) and SAS 9.1 (SAS Institute Inc., USA) were used for analyses.
|
Results |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsDiscussionFundingAcknowledgementReference |
|---|
Baseline characteristics
Demographics, clinical characteristics and endocrine profile on c.d. 2 are shown in Table I. Although properly randomized, the proportion of male factor infertility was higher in the priming compared to control group, 34/53 versus 19/50, respectively (P = 0.008). Otherwise groups were comparable.
|
Primary endpoint
The number of top-quality embryos was 1.08 (0.83–1.40) and 1.43 (1.12–1.81) in the priming and control group, respectively. When the number of top-quality embryos was treated as a Poisson distributed count, and the mean values were compared for the two groups, the control group had an estimated 32% higher mean count than the priming group (95% CI from 7% below to 89% above), but the difference was not significant (P = 0.120), Table II.
|
Secondary endpoints
Stimulation characteristics
Duration of ovarian stimulation and rFSH dose
The number of stimulation days required to reach the criterion of GnRH antagonist administration or hCG administration was significantly higher in the priming group compared to control group (P < 0.001), as were the total dose of rFSH administered (P < 0.001), Table III. However, the mean rFSH dose per stimulation day (median 183 IU (173–188) versus 175 IU (163–178), P = 0.068) and the proportion of women with dose increment (75 IU) on stimulation day 6 (31/53 versus 26/50, P = 0.508), on stimulation day 8 (14/51 versus 10/36, P = 0.973) and on stimulation day 10 (4/29 versus 1/9, P = 0.835) was similar in the priming and control group, respectively. Only one patient in each group had a dose reduction during treatment.
|
Follicular dynamics
Within the priming group, there were no statistically significant changes in number and size of follicles during the priming period from c.d. 2 to 5. During the first 6 days of ovarian stimulation, the follicular growth was slower in the priming group compared to control group with number of follicles >10 mm being lower on stimulation day 6 (P < 0.001) (Table III). On the day of hCG, the total number of follicles was same in the two groups as was the number of follicles according to different size categories. However, the proportion of c.d. 2 antral follicles that reached the sizes >10,
12 or 14 mm on day of hCG was significantly higher in the priming group (Table III).
Endometrium
Endometrial thickness was similar in the two groups at baseline (c.d. 2). During the priming period from c.d. 2 to 5, the endometrial thickness decreased significantly in the priming group (P < 0.001), leading to a significantly thinner endometrium in the priming group compared to control group on stimulation day 1. However, this was reversed during ovarian stimulation (Table III).
Cycle outcome
Cycle outcome is shown in Table II. The number of oocytes retrieved, the number and quality of embryos, the implantation rate, pregnancy rate and the delivery rate did not differ significantly between groups. The fertilization rate was significantly lower in the priming group (P = 0.007), also when adjusted for fertilization method used (P = 0.031). A total of 34 healthy children were born including three pairs of twins. Single embryo transfer was performed in 56% of the transfers (53/94).
Serum endocrinology
Serum steroid levels during treatment are shown in Fig. 3. On stimulation day 1, the levels of gonadotrophins were significantly lower in the priming group that had received GnRH antagonist during the priming period as compared to the control group. Within the priming group, the E2 level decreased during the priming period from 0.13 nmol/l (median, 95% CI 0.12–0.15) on c.d. 2 to 0.09 nmol/l (0.07–0.10) on c.d. 5 (stimulation day 1) (P < 0.001). The E2 level remained significantly lower in the priming versus control group during the first 6 days of stimulation (P 0.004). In contrast, on the day of hCG administration, the level of E2 was higher in the priming group (P < 0.001), as was the level of E2 per follicle >10 mm (P = 0.011), per follicle 12 mm (P = 0.013) and per follicle 14 mm (P = 0.005) (Table III). The testosterone level did not differ significantly between the two groups at any measured time point except on the day of hCG administration, where the level was 1.5 nmol/l (1.3–1.7) in the priming group and 1.2 nmol/l (1.1–1.5) in the control group (P = 0.041). The progesterone level did not differ significantly between the two groups at any measured time point (Fig. 3). Nevertheless, within the priming group, the progesterone level significantly increased from c.d. 2 to 5 (P = 0.014), and significantly decreased from c.d. 5 to 7 (P < 0.001), indicating some short-term drug induced reactivation of corpus luteum after administration of a bolus of 1250 IU hCG on c.d. 2. In the control group, no subject (0/50) had a progesterone rise of more than 100% from c.d. 2 to 4 (stimulation day 1–3) as compared to the priming group, where 22.6% (12/53) had a progesterone rise of more than 100% during the priming period from c.d. 2 to 5 (P < 0.001).
|
Administration of 1250 IU hCG on c.d. 2 in the priming group led to the following serum hCG levels (IU/l): 12.3 (95% CI 11.1–13.4) on c.d. 5 (stimulation day 1), 4.5 (95% CI 3.9–5.1) on stimulation day 3 and below 2.0 in all women on stimulation day 6.
Premature rise of LH
One patient in each of the priming and control groups experienced a premature LH rise to 20.1 and 22.2 IU/l, respectively. Both rises occurred on stimulation day 8 before the onset of daily GnRH antagonist administration according to follicular size. The two women continued treatment, resulting in an ongoing pregnancy in the woman from the priming group.
FF analyses
FF hormone levels are shown in Table IV. The testosterone (P = 0.014) and the E2 levels (P = 0.007) were significantly higher in the priming group, where patients were treated with aromatase inhibitor and hCG during early follicular phase GnRH antagonist administration, as compared to the control group. The androstendione level was also higher in the priming group, but this finding did not reach statistical significance (P = 0.070). Levels of progesterone, inhibin B and AMH did not differ statistically between the two groups, nor did the ratios of E2 to testosterone, androstendione or progesterone.
Associations between hormonal levels in the pre-ovulatory follicles are shown in Table V. Highly significant positive correlations were found between FF androgen levels and E2 (P < 0.001), whereas significant negative correlations were found between FF androgen levels and inhibin B (P < 0.001). Further, a weak but significant negative correlation was found between the progesterone level and the level of both AMH (P < 0.001) and inhibin B (P = 0.018).
|
Adverse reactions/events
No serious adverse reactions or events occurred in any of the groups. A total of 34 healthy children were born as a result of treatment. No malformations were recorded.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsDiscussionFundingAcknowledgementReference |
|---|
In the present study, we tested whether a short-term androgen priming by aromatase inhibitor and hCG before COS increases the number of top-quality embryos after IVF/ICSI. The number of embryos transferred is declining, and often restricted to one or two to limit the number of multiple gestations, which makes the quality of the transferred embryos highly important (Ziebe et al., 1997
). The number of top-quality embryos was not increased by the introduction of a 3-day period of anastrozole and hCG priming during early-follicular-phase GnRH antagonist administration when compared to a standard antagonist protocol.
In our previous non-randomized pilot study, significantly more embryos and top-quality embryos were found in the priming group that received letrozole and hCG during a 7-day priming period (Lossl et al., 2006; Andersen and Lossl, 2007). However, the pre-ovulatory FF, obtained almost 2 weeks after cessation of the priming period, showed an increased testosterone level and decreases in both the E2 to testosterone ratio and the inhibin B level in the priming group as compared to the control group. Previous studies showed that a higher ratio of E2 to androgen in pre-ovulatory FFs reflected both good follicular health and the viability of the enclosed oocyte (McNatty et al., 1979; Andersen, 1993), and that the pre-ovulatory FF level of inhibin B was positively associated with embryo score (Chang et al., 2002). This suggests that the granulosa cells of the ‘androgen-primed’ follicles functioned sub-optimally, and that the aromatase inhibitor was perhaps not fully cleared from the follicular environment. The present study was designed both with this in mind and to simplify the protocol. We therefore decided to (i) use anastrozole that is a less potent aromatase inhibitor than letrozole (Lonning et al., 2003b), (ii) use a lower dosage of aromatase inhibitor, 1.0 mg daily instead of 2.5 mg daily and (iii) shorten the priming period from 7 to 3 days. The pre-ovulatory FF level of testosterone remained slightly but significantly increased in the priming group compared to the control group. However, the E2 level was also increased, and the E2 to androgen ratio and the level of inhibin B were now similar in the priming and control groups. The clinical relevance of these reported differences in FF hormonal concentrations remains unknown. Certainly, they were not translated into better cycle outcome in the present study, as the number of top-quality embryos, the implantation rate, the pregnancy rate and the delivery rate was not improved in the priming group compared to the control group. The present study was not sufficiently powered to detect smaller differences in pregnancy and live birth rates between groups, and for this reason they cannot be excluded. Indeed, the fertilization rate was lower in the priming group, suggesting that priming may have impaired oocyte quality. It is uncertain whether the change of protocol or the much larger sample size in the present study were responsible for the differences between the results of the preliminary and the present study. However, any clinical benefits of short-term androgen priming by use of aromatase inhibitor and hCG remain unproven.
We are unable to conclude whether the intraovarian androgen level was genuinely increased during the priming period and whether the granulosa cell FSH receptor level was enhanced. Anastrozole and hCG have half-lives of 41 and 32 hours, respectively (Mannaerts et al., 1998; Lonning et al., 2003a), implying that only small and negligible concentrations of these drugs persist in circulation in the late follicular phase. Thus, the higher testosterone level in pre-ovulatory FF in the priming group compared to the control group indicates that follicles from women in the priming group may have experienced enhanced levels of androgens during the entire length of the follicular phase. We cannot exclude the possibility that the longer duration of rFSH stimulation seen in the priming group, or the minor dose adjustments allowed from stimulation day 6, may have affected the intrafollicular steroid levels, despite the fact that the criterion for hCG administration, the mean daily rFSH dose, and the proportion of patients with dose increments were similar in the two groups.
The proportion of c.d. 2 antral follicles that had reached a diameter >10, 12 or 14 mm by the day of hCG administration was significantly higher in the priming group compared to the control group suggesting an enhanced secondary recruitment. On the day of hCG administration, the serum E2 level and the serum E2 level per follicle >10, 12 or 14 mm were increased in the priming group, indicating a higher level of granulosa cell aromatase activity. Increased secondary recruitment and increased aromatization could be explained by a higher follicular FSH (LH) sensitivity, as serum levels of gonadotropins were similar in the priming and control groups during the stimulation.
Interestingly, short-term androgen priming by aromatase inhibitor and hCG before COS caused changes in follicular growth and hormone profiles that were quite similar to those seen after stimulation with hCG-containing gonadotrophins in the long agonist protocol (Andersen et al., 2006; Smitz et al., 2007). The duration of stimulation was longer, the number of follicles >10 mm on stimulation day 6 were fewer, serum E2 level was lower on stimulation day 6 but higher on the day of hCG, and pre-ovulatory FF levels of E2 and androgens were higher in the presence of LH activity. The authors suggested that paracrine factors induced by hCG, and mediated by androgens, activated the aromatase system and caused a more estrogenic intrafollicular environment by the end of stimulation.
In the present study, a drug induced serum progesterone rise was seen during the priming period caused by a bolus of 1250 IU hCG administered on c.d. 2. This iatrogenic increase seemed fully compatible with pregnancy, as the ongoing pregnancy rate was 33% (4/12) among the patients with the largest relative progesterone rise (>100%) during the priming period.
The negative correlations of FF androgens and inhibin B in the present study again suggest a specific negative effect of androgens on granulosa cell inhibin B synthesis as hypothesized in our preliminary study (Andersen and Lossl, 2007).
We have studied a ‘local’ approach of androgen priming in ‘normal responders’ for COS including an aromatase inhibitor. The safety of aromatase inhibitors in assisted reproduction has been raised, as a higher prevalence of locomotor abnormalities was found among 150 children born after the use of letrozole for ovulation induction when compared to 36 000 children born after spontaneous conception in a non-infertile, low risk population of women (Biljan et al., 2005). However, the study group was much smaller, older, had a higher multiple birth rate and a high rate of gestational diabetes, which may have biased the results. A further study comparing 514 children born after ovarian stimulation including letrozole with 397 children born after use of chlomiphene citrate (Tulandi et al., 2006) found no increase in the malformation rate when letrozole was administered (1.2 versus 3.0%).
A more ‘systemic’ approach, with administration of testosterone or dehydroepiandrosterone (DHEA) before COS, has been studied in ‘poor responder patients’. When comparing a poor response cycle with a following experimental cycle an increase in the number of mature follicles from 1.0 to 2.2 was reported in five women who received 8 weeks of DHEA priming (Casson et al., 2000), and an increase in the number of oocytes retrieved from mean (SEM) 3.4 ± 0.5 to 4.4 ± 0.5 was reported in 25 women who received a mean of 17.6 weeks of DHEA priming (Barad and Gleicher, 2006). However, these improvements do not seem to differ from the improvements found in the placebo group of the only randomized controlled trial on ‘systemic’ androgen priming (Massin et al., 2006), where the number of oocytes retrieved increased from 3.6 ± 1.1 to 5.0 ± 2.1, demonstrating the ‘regression towards the mean’ with a more normal response in the subsequent cycle. Patients received testosterone gel (n = 24) or placebo (n = 25) for 15 days before COS and no significant differences in the number of pre-ovulatory follicles, oocytes retrieved or embryos were found between the groups. On the other hand Balasch et al. (2006) found an increase in the number of follicles >10 mm from 1.6 ± 0.2 (first cycle) and 1.6 ± 0.3 (second cycle) to 8.5 ± 0.7 after only 5 days of testosterone priming in 20 patients with two earlier ‘poor response cycles’. Finally an improvement in ovarian response to COS over nine successive cycles was reported in a 43-year-old woman after self-administering daily DHEA (Barad and Gleicher, 2005). These studies confirm that more research is needed before firm conclusions can be drawn as to the clinical relevance of androgen priming.
To conclude, the use of aromatase inhibitor and hCG during early-follicular-phase GnRH antagonist administration before COS did not increase the number of top-quality embryos when compared to a standard antagonist protocol. The use of aromatase inhibitor and hCG seems to be a valid method of increasing the local intraovarian androgen level compatible with achievement of pregnancy, but the concept of androgen priming needs further optimization to elucidate possible clinical use.
|
Funding |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsDiscussionFundingAcknowledgementReference |
|---|
Merck Serono is thanked for an unrestricted educational grant supporting Kristine Lossl.
|
Acknowledgement |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsDiscussionFundingAcknowledgementReference |
|---|
We thank Associate Professor Lene Theil Skovgaard, Department of Biostatistics, University of Copenhagen, Denmark for statistical support. ClinicalTrials.gov Identifier: NCT00286364 [ClinicalTrials.gov] .
|
Reference |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsDiscussionFundingAcknowledgementReference |
|---|
Andersen CY. Characteristics of human follicular fluid associated with successful conception after in vitro fertilization. J Clin Endocrinol Metab (1993) 77:1227–1234.[Abstract]
Andersen CY, Lossl K. Increased intrafollicular androgen levels affect human granulosa cell secretion of anti-mullerian hormone and inhibin-B. Fertil Steril (2007) in press.
Andersen AN, Devroey P, Arce JC. Clinical outcome following stimulation with highly purified hMG or recombinant FSH in patients undergoing IVF: a randomized assessor-blind controlled trial. Hum Reprod (2006) 21:3217–3227.
Balasch J, Fabregues F, Penarrubia J, Carmona F, Casamitjana R, Creus M, Manau D, Casals G, Vanrell JA. Pretreatment with transdermal testosterone may improve ovarian response to gonadotrophins in poor-responder IVF patients with normal basal concentrations of FSH. Hum Reprod (2006) 21:1884–1893.
Barad DH, Gleicher N. Increased oocyte production after treatment with dehydroepiandrosterone. Fertil Steril (2005) 84:756.[Medline]
Barad D, Gleicher N. Effect of dehydroepiandrosterone on oocyte and embryo yields, embryo grade and cell number in IVF. Hum Reprod (2006) 21:2845–2849.
Biljan M, Hemming R, Brassard N. The outcome of 150 babies following the treatment with letrozole or letrozole and gonadotropins. Fertil Steril (2005) 84(Suppl 1):S95.
Casson PR, Lindsay MS, Pisarska MD, Carson SA, Buster JE. Dehydroepiandrosterone supplementation augments ovarian stimulation in poor responders: a case series. Hum Reprod (2000) 15:2129–2132.
Chang CL, Wang TH, Horng SG, Wu HM, Wang HS, Soong YK. The concentration of inhibin B in follicular fluid: relation to oocyte maturation and embryo development. Hum Reprod (2002) 17:1724–1728.
Lonning P, Pfister C, Martoni A, Zamagni C. Pharmacokinetics of third-generation aromatase inhibitors. Semin Oncol (2003) a 30:23–32.[Medline]
Lonning PE, Geisler J, Bhatnager A. Development of aromatase inhibitors and their pharmacologic profile. Am J Clin Oncol (2003) b 26:S3–S8.[CrossRef][Web of Science][Medline]
Lossl K, Andersen AN, Loft A, Freiesleben NL, Bangsboll S, Andersen CY. Androgen priming using aromatase inhibitor and hCG during early-follicular-phase GnRH antagonist down-regulation in modified antagonist protocols. Hum Reprod (2006) 21:2593–2600.
Luo W, Wiltbank MC. Distinct regulation by steroids of messenger RNAs for FSHR and CYP19A1 in bovine granulosa cells. Biol Reprod (2006) 75:217–225.
Mannaerts BM, Geurts TB, Odink J. A randomized three-way cross-over study in healthy pituitary-suppressed women to compare the bioavailability of human chorionic gonadotrophin (Pregnyl) after intramuscular and subcutaneous administration. Hum Reprod (1998) 13:1461–1464.
Massin N, Cedrin-Durnerin I, Coussieu C, Galey-Fontaine J, Wolf JP, Hugues JN. Effects of transdermal testosterone application on the ovarian response to FSH in poor responders undergoing assisted reproduction technique–a prospective, randomized, double-blind study. Hum Reprod (2006) 21:1204–1211.
McNatty KP, Smith DM, Makris A, Osathanondh R, Ryan KJ. The microenvironment of the human antral follicle: interrelationships among the steroid levels in antral fluid, the population of granulosa cells, and the status of the oocyte in vivo and in vitro. J Clin Endocrinol Metab (1979) 49:851–860.
Mitwally MF, Casper RF. Use of an aromatase inhibitor for induction of ovulation in patients with an inadequate response to clomiphene citrate. Fertil Steril (2001) 75:305–309.[CrossRef][Web of Science][Medline]
Rani CS, Salhanick AR, Armstrong DT. Follicle-stimulating hormone induction of luteinizing hormone receptor in cultured rat granulosa cells: an examination of the need for steroids in the induction process. Endocrinology (1981) 108:1379–1385.
Shaw HJ, Hillier SG, Hodges JK. Developmental changes in luteinizing hormone/human chorionic gonadotropin steroidogenic responsiveness in marmoset granulosa cells: effects of follicle-stimulating hormone and androgens. Endocrinology (1989) 124:1669–1677.
Smitz J, Andersen AN, Devroey P, Arce JC. Endocrine profile in serum and follicular fluid differs after ovarian stimulation with HP-hMG or recombinant FSH in IVF patients. Hum Reprod (2007) 22:676–687.
The Rotterdam ESHRE/ASRM-Sponsered PCOS consensus workshop group. Revised 2003 consensus on diagnostic criteria and long-term health risks related to polycystic ovary syndrome (PCOS). Hum Reprod (2004) 19:41–47.
Tulandi T, Martin J, Al-Fadhli R, Kabli N, Forman R, Hitkari J, Librach C, Greenblatt E, Casper RF. Congenital malformations among 911 newborns conceived after infertility treatment with letrozole or clomiphene citrate. Fertil Steril (2006) 85:1761–1765.[CrossRef][Web of Science][Medline]
Vendola KA, Zhou J, Adesanya OO, Weil SJ, Bondy CA. Androgens stimulate early stages of follicular growth in the primate ovary. J Clin Invest (1998) 101:2622–2629.[Web of Science][Medline]
Weil S, Vendola K, Zhou J, Bondy CA. Androgen and follicle-stimulating hormone interactions in primate ovarian follicle development. J Clin Endocrinol Metab (1999) 84:2951–2956.
Ziebe S, Petersen K, Lindenberg S, Andersen AG, Gabrielsen A, Andersen AN. Embryo morphology or cleavage stage: how to select the best embryos for transfer after in-vitro fertilization. Hum Reprod (1997) 12:1545–1549.
Submitted on November 7, 2007; resubmitted on March 14, 2008; accepted on March 28, 2008.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  J. Balasch, F. Fabregues, F. Carmona, R. Casamitjana, and M. Tena-Sempere Ovarian Luteinizing Hormone Priming Preceding Follicle-Stimulating Hormone Stimulation: Clinical and Endocrine Effects in Women with Long-Term Hypogonadotropic Hypogonadism J. Clin. Endocrinol. Metab., July 1, 2009; 94(7): 2367 - 2373. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Fabregues, J. Penarrubia, M. Creus, D. Manau, G. Casals, F. Carmona, and J. Balasch Transdermal testosterone may improve ovarian response to gonadotrophins in low-responder IVF patients: a randomized, clinical trial Hum. Reprod., February 1, 2009; 24(2): 349 - 359. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Lossl, C. Yding Andersen, A. Loft, and A. Nyboe Andersen Reply: Androgen priming before ovarian stimulation for IVF Hum. Reprod., December 1, 2008; 23(12): 2870 - 2871. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||