Abstract. RBMOnline - Vol 7. No Reproductive BioMedicine Online; on web 20 May 2003

Transcription

1 RBMOnline - Vol 7. No Reproductive BioMedicine Online; on web 20 May 2003 Article Ovarian responses to recombinant FSH or HMG in normogonadotrophic women following pituitary desensitization by a depot GnRH agonist for assisted reproduction Dr Juan Balasch Juan Balasch obtained his MD degree (1974) and the speciality degree in Obstetrics and Gynaecology (1977) at the Faculty of Medicine-Hospital Clínic, University of Barcelona in Spain. The PhD degree was granted to him at the same University in At present he is full professor in Obstetrics and Gynaecology and Head of the Fertility Unit at the Faculty of Medicine-Hospital Clinic, University of Barcelona. Professor Balasch is Past President of the Spanish Fertility Society and has more than 150 publications in international journals and books; in a series of studies he and his team developed a new hypothesis on the pathogenesis of the ovarian hyperstimulation syndrome. He serves as ad-hoc reviewer or is in the Editorial Board of different international journals dealing with fertility, gynaecological endocrinology and human reproduction. Professor Balasch s current research interests include assisted reproduction, repeated abortion, implantation failure and ovarian (hyper)stimulation. Juan Balasch 1,2, Joana Peñarrubia 1, Francisco Fábregues 1, Ester Vidal 1, Roser Casamitjana 3, Dolors Manau 1, Francisco Carmona 1, Montserrat Creus 1, Juan A Vanrell 1 Institut Clínic of Gynecology, Obstetrics and Neonatology, and Hormonal Laboratory, Faculty of Medicine, University of Barcelona, Hospital Clínic, Institut d Investigacions Biomèdiques August Pi i Sunyer (IDIBAPS), Barcelona, Spain 2 Correspondence: Fax: ; jbalasch@medicina.ub.es Abstract At present, there is considerable debate about the utility of supplemental LH in assisted reproduction treatment. In order to explore this, the present authors used a depot gonadotrophin-releasing hormone agonist (GnRHa) protocol combined with recombinant human FSH (rhfsh) or human menopausal gonadotrophin (HMG) in patients undergoing intracytoplasmic sperm injection (ICSI). The response to either rhfsh (75 IU FSH/ampoule; group rhfsh, 25 patients) or HMG (75 IU FSH and 75 IU LH/ampoule; group HMG, 25 patients) was compared in normo-ovulatory women suppressed with a depot triptorelin injection and candidates for ICSI. A fixed regimen of 150 IU rhfsh or HMG was administered in the first 14 days of treatment. Treatment was monitored with transvaginal pelvic ultrasonographic scans and serum measurement of FSH, LH, oestradiol, androstenedione, testosterone, progesterone, inhibin A, inhibin B and human chorionic gonadotrophin (HCG) at 2-day intervals. Although oestradiol serum concentrations on the day of HCG injection were similar, both the duration of treatment and the per cycle gonadotrophin dose were lower in group HMG. In the initial 16 days of gonadotrophin treatment, the area under the curve (AUC) of LH, oestradiol, androstenedione and inhibin B were higher in group HMG; no differences were seen for the remaining hormones measured, including the inhibin B:inhibin A ratio. The dynamics of ovarian follicle development during gonadotrophin treatment were similar in both study groups, but there were more leading follicles (>17 mm in diameter) on the day of HCG injection in the rhfsh group. The number of oocytes, mature oocytes and good quality zygotes and embryos obtained were significantly increased in the rhfsh group. It is concluded that in IVF patients undergoing pituitary desensitization with a depot agonist preparation, supplemental LH may be required in terms of treatment duration and gonadotrophin consumption. However, both oocyte, embryo yield and quality were significantly higher with the use of rhfsh. Keywords: HMG, IVF, ovarian stimulation, pituitary suppression, recombinant FSH 35

2 36 Introduction The optimal ratio of FSH to LH activity in the stimulating drugs used for ovulation induction has been a matter of debate since the early days of gonadotrophin therapy (Jacobson and Marshall, 1969; Louwerens, 1969). At present, there is a renewed interest in the subject related to the use of recombinant human FSH (rhfsh) preparations completely devoid of LH activity for controlled ovarian stimulation to induce multiple follicular development in assisted reproduction cycles. Thus, it is well established that successful IVF and embryo transfer requires both stimulation of the ovary and suppression of the pituitary. Exogenous gonadotrophins and gonadotrophin-releasing hormone (GnRH) analogues have been the key hormones required to maximize IVF success (Barbieri and Hornstein, 1999; Ludwig et al., 2002; Cohen, 2003). However, the low endogenous LH concentrations achieved with GnRH agonists may amplify the differences, if any, in treatment outcomes seen with the use of human menopausal gonadotrophin (HMG) and FSH-only preparations. This is especially true with the use of the longacting depot preparations, which are convenient for patient use but have potential disadvantages because they produce more profound pituitary desensitization than do the short-acting preparations (Yim et al., 2001). On the above evidence, a recent clinical trial (Filicori et al., 2001) compared the ovarian responses to either highly purified FSH or HMG in 50 normo-ovulatory pituitary-suppressed women by a single depot GnRH-agonist injection in candidates for intrauterine insemination (IUI). From that study it was concluded that ovulation induction with LH activitycontaining menotropins is associated with shorter treatment duration, lower menotropin consumption and reduced development of small ovarian follicles (Filicori et al., 2001). In that previous study (Filicori et al., 2001), however, only IUI patients were included and, therefore, information regarding the number, maturity and developmental competence of oocytes obtained as a consequence of ovarian stimulation with two different gonadotrophin preparations was not provided; thus a meaningful piece of information was lost. The present prospective, randomized clinical trial was set up in patients undergoing intracytoplasmic sperm injection (ICSI) procedures to provide additional information on the comparative effect of both gonadotrophins on oocyte number and quality. Thus, patient population characteristics, ovarian stimulation protocol and monitoring and hormone analyses were the same as those in the previous report (Filicori et al., 2001). Materials and methods Patients, design and sample size Fifty patients from an ICSI programme who had unexplained or male-related primary infertility were included in the present study, which was approved by the clinic s internal ethical committee. All patients provided informed consent. All the subjects were regularly menstruating (menstrual cycles of days) premenopausal women aged years and having a normal body mass index (BMI) of kg/m 2. All the women had both ovaries and no occult ovarian failure on the basis of their cycle day 3 FSH concentration of <10 IU/l (range IU/l) (standard International Reference Preparation (IRP) 78/549) measured in the cycle preceding ICSI. The subjects had undergone a complete infertility evaluation, including laparoscopy and ultrasound examination of the ovaries. None of the women had polycystic ovarian disease (PCOD), according to ultrasonographic or basal gonadotrophin measurement. Couples with unexplained infertility had been treated with ovulation induction and IUI for at least three cycles but had failed to conceive. No patient had received any hormone therapy, including gonadotrophins, for at least 6 months preceding the study. All patients included underwent their first assisted reproduction treatment cycle, and in all of them ICSI was used, thus allowing the estimation of oocyte maturity and developmental potential directly at the time of recovery. Moreover, the conditions of ICSI restrict the multifactorial nature of fertilization success and failure, making prominent those factors that are responsible for oocyte activation and the ensuing developmental processes culminating in the completion of oocyte meiosis and in the development of pronuclei (Mendoza et al., 1999). Furthermore, in this study, matched pairs were randomized to balance other baseline potentially confounding variables of ovarian response to gonadotrophins and IVF/ICSI, such as age of the woman (±1 year), indication for ICSI (male factor or unexplained infertility), and BMI (±1 kg/m 2 ). Initially, 60 patients participating in the ICSI programme were selected to obtain pairs of women matched on these IVF/ICSI factors. Members of each pair were then allocated randomly (according to a computer-generated table) to receive rhfsh (group rhfsh, n = 30) or HMG (group HMG, n = 30). The sample size (25 patients per group) was the same as in the previous study on the subject (Filicori et al., 2001). However, initially more pairs of patients were included to allow for cancelled cycles, to arrive at the final sample of at least 25 patients per group. Among the 60 women initially selected, several had cancelled cycles due to a low response; 25 pairs of patients undergoing follicular aspiration were finally considered in the evaluation of the results. Cancellation rates were 13.3% (4 patients) and 20% (6 patients) in groups rhfsh and HMG, respectively. Stimulation regimen As in the previous report (Filicori et al., 2001), treatment was started in the mid-luteal phase of a spontaneous menstrual cycle with the administration of a single injection of 3.75 mg depot triptorelin (Decapeptyl 3.75, Ipsen-Pharma, Barcelona, Spain). Ovarian stimulation with gonadotrophins was commenced days thereafter when serum oestradiol levels declined to <50 pg/ml and a vaginal ultrasonographic scan showed an absence of follicles >10 mm in diameter. Patients in group rhfsh received two ampoules (150 IU) s.c. daily of rhfsh (Gonal-F, Serono, Madrid, Spain) and patients in group HMG received two ampoules i.m. daily of HMG (150 IU FSH and 150 IU LH; Pergonal, Serono). In all patients, gonadotrophins were administered at 16:00 18:00 hours, and in both groups the gonadotrophin administration schedule was not changed for 14 days or until at least four ovarian follicles of 14 mm in diameter with the two leading follicles measuring >17 mm in association with a

3 consistent rise in serum oestradiol concentration were detected (final maturation parameters). If these parameters were not achieved by the 14th day of treatment, increments in the rhfsh and HMG doses were allowed, according to the individual ovarian response, with the following schedule: three daily ampoules of rhfsh or HMG on stimulation days 15 17, and four daily ampoules of rhfsh or HMG thereafter. At obtainment of the final maturation parameters, 10,000 IU HCG (Profasi, Serono) were administered to induce final follicular maturation and early luteinization. Oocyte aspiration was performed with vaginal ultrasonography h after HCG administration. The maturational status of the oocytes and embryo grading were recorded according to published criteria (Veeck, 1999); embryos of Veeck grades 1 or 2 were considered high quality. ICSI was performed on metaphase II oocytes only. The embryologists performing the assessment of oocytes and ICSI were blinded to the type of gonadotrophin each patient had received for stimulation. Two days later, two to three embryos per patient were replaced, depending on the age of the patient, the indication for ICSI and the number and quality of embryos available for replacement. Two additional doses of 2500 IU of HCG (Profasi, Serono) on the days of embryo transfer and 3 days later were given to supplement the luteal phase in all patients. Pregnancy was diagnosed by increasing serum concentrations of β-hcg after embryo transfer, and the subsequent demonstration of an intrauterine gestational sac by ultrasonography. Monitoring, hormone analyses and ultrasonography Treatment monitoring was carried out throughout gonadotrophin administration. Blood samples for hormone analyses were obtained on day 0 (day of pituitary suppression) and every other day from stimulation day 6 until the day of HCG injection. Ultrasonographic scans were performed on the same days as hormone analyses. Each day one blood sample was drawn between 08:00 and 10:00 hours, and for this study two serum aliquots were obtained. Oestradiol was measured daily in one of the serum aliquots for clinical monitoring, and the second aliquot was stored at 20 C for later measurements of FSH, LH, androstenedione, testosterone, progesterone, inhibin A, inhibin B and HCG. Frozen blood samples from each patient were examined in one run. In no case did the ultrasonographer or the hormonal laboratory know the treatment groups in which the patients were included. Hormones were measured using commercially available kits, as reported previously (Balasch et al., 1998, 2001; Peñarrubia et al., 2000; Creus et al., 2001). FSH, LH and prolactin serum concentrations were measured by an immunoenzymatic assay with two monoclonal antibodies (Immuno 1, Technicon; Bayer, Tarrytown, NY, USA). Data are expressed in terms of IRP 78/549 and 68/40 for FSH and LH, respectively. The sensitivity of the assays was 0.1 IU/l for FSH, 0.3 IU/l for LH, and 0.1 ng/ml for prolactin, and the interassay coefficients of variation (CV) were 2.7, 3.1 and 4.4% respectively. Oestradiol, testosterone and progesterone concentrations in serum were estimated by a competitive immunoenzymatic assay (Immuno 1, Technicon; Bayer). The sensitivity was 10 pg/ml for oestradiol, 8 ng/dl for testosterone and 0.2 ng/ml for progesterone, and the interassay CV were 5, 9.5 and 6.7%, respectively. Androstenedione in serum was measured by a competitive radioimmunoassay (Diagnostic Systems Laboratories, Webster, Texas, USA); the sensitivity of the assay was 10 ng/dl, and the interassay CV was 12%. Inhibin A and inhibin B measurements were performed by a solid-phase sandwich enzyme-linked immunosorbent assay (ELISA) carried out in microtitre plates (Serotec, Oxford, UK). The sensitivity of the assay was 2 pg/ml for inhibin A and 15 pg/ml for inhibin B. The interassay CV for inhibin A at low (21 pg/ml) and high (168 pg/ml) concentrations were 15 and 9%, respectively. The interassay CV for inhibin B at low (36 pg/ml) and high (246 pg/ml) concentrations were 12 and 7%, respectively. Total β-hcg was measured by a solid-phase two-site chemiluminescent enzyme immunometric assay standardized against the Third International Standard 75/537 (Immulite; Diagnostic Products, Los Angeles, CA, USA) with a detection limit of 2 IU/l. The interassay CV was 5.8%. Ultrasonic scans were performed using a 5 mhz vaginal transducer attached to an Aloka sector scanner (Model SSD- 620; Aloka, Tokyo, Japan). Statistics Data were analysed by Statistics Package for Social Sciences statistical software using the Wilcoxon matched-pairs signedranks test and the chi-squared test, as appropriate. Serum hormone concentrations during the first 16 days of treatment were calculated in each cycle as the area under the curve (AUC) which corresponds to an integrated hormone concentration for the chosen days. Integrated concentrations were compared between treatment groups by the Wilcoxon matched-pairs signed-ranks test. Similarly, the dynamics of ovarian follicle development were compared between treatment groups for the first 16 days of gonadotrophin therapy. Results are expressed as means with SE. Results The results are summarized in Tables 1 3 and Figures 1 and 2. As shown in Table 1, the main demographic and baseline characteristics of the patients in groups rhfsh and HMG, including age, BMI, duration of infertility and hormone measurements on cycle day 3, were almost identical. As expected, the causes of infertility were identical for both groups. This supports the validity of the randomization process. Table 2 shows data regarding gonadotrophin treatment, ovarian response, oocyte retrieval and ICSI outcome in the two groups studied. Both the duration of ovarian stimulation and the total number of ampoules of gonadotrophin administered were significantly increased in the rhfsh group. The serum concentration of oestradiol and the endometrial thickness on the day of HCG injection were similar for the two groups of ICSI patients. However, the number of leading follicles (>17 mm) on the day of HCG injection was significantly higher in group rhfsh. There were no differences between groups with respect to the number of small (<10 mm), medium size (10 14 mm) and large (>14 17 mm) ovarian follicles 37

4 Table 1. Patient baseline characteristics in groups rhfsh and HMG. Values are means ± SE unless otherwise stated. Variable rhfsh group HMG group (n = 25) (n = 25) P-value Age (years) ± ± 0.47 NS BMI (kg/m 2 ) ± ± 0.53 NS Infertility factor Male factor (n) 17 (68) 17 (68) Unexplained (n) 8 (32) 8 (32) NS Duration of infertility 5.31 ± ± 0.43 NS (years) Day 3 FSH (IU/l) 7.49 ± ± 0.22 NS Day 3 LH (IU/l) 5.97 ± ± 0.70 NS Day 3 prolactin (ng/ml) ± ± 8.40 NS Day 3 oestradiol (pg/ml) ± ± 4.34 NS Day 3 inhibin B (pg/ml) ± ± 4.10 NS Values in parentheses are percentages. rhfsh = recombinant human FSH; HMG = human menopausal gonadotrophin; BMI = body mass index; NS = not significant. detected on the day of HCG administration. 38 Duration of treatment ranged between days and 8 16 days in groups rhfsh and HMG, respectively. There were still three HMG patients on treatment on day 16, and four rhfsh patients on day 18. Three patients in group HMG had to undergo one increment of the HMG dose, while eight patients in group rhfsh underwent one gonadotrophin dose increment, and five of them underwent a second increment. Thus, follicular dynamics and hormone concentrations during ovarian stimulation with the two regimens of gonadotrophins were compared for the first 16 days of treatment. The dynamics of ovarian follicle development during gonadotrophin treatment in both study groups are presented in Figure 1. The number of small, medium and large ovarian follicles measured at 2-day intervals did not differ throughout treatment between groups rhfsh and HMG. Figure 2 shows the follicular phase secretory dynamics of daily gonadotrophin, ovarian steroids, and inhibin A and inhibin B levels throughout the treatment period. Hormone concentrations calculated as the AUC are presented in Table 3. Serum concentrations of FSH, testosterone, progesterone and inhibin A, and the inhibin B:inhibin A ratio, did not differ between treatment groups. However, the AUC for LH, oestradiol, and inhibin B was significantly higher in group HMG. The AUC for androstenedione was also higher in group HMG compared with that in group rhfsh, the difference almost reaching statistical significance (P < 0.06). Serum levels of β-hcg were below the detection limit of the assay throughout the study period in both groups of patients (data not shown). All patients underwent HCG injection and oocyte retrieval, but the oocyte yield, the number of metaphase II oocytes, the fertilization rate and the number of viable and high quality embryos were significantly higher in group rhfsh (Table 2). Implantation and pregnancy rates were similar in groups rhfsh and HMG. There were three pairs of twins in group rhfsh and four pairs of twins in the HMG group. No patient Figure 1. Number (mean ± SE) of small (<10 mm in diameter), medium (10 14 mm), and large (>14 mm) ovarian follicles measured with transvaginal ultrasonography during gonadotrophin treatment at 2-day intervals in the recombinant human FSH (rhfsh) group and the human menopausal gonadotrophin (HMG) group. No differences were found between groups at any study point. There were no patients on treatment beyond day 16 in the HMG group. developed ovarian hyperstimulation syndrome (OHSS), and there were one (11.1%) and three (33.3%) miscarriages in groups rhfsh and HMG, respectively. Discussion This study employed a fixed low gonadotrophin dose and the use of a depot agonist preparation (both of which are neither standard practices nor absolutely first choice in assisted reproduction treatment), features that were used in a previous study including IUI patients (Filicori et al., 2001). This was done in order to examine the effect of profound LH suppression in patients undergoing ICSI, thus in principle favouring HMG. Whether women undergoing ovulation induction protocols with pituitary desensitization using depot agonists would need exogenous LH in order to modulate treatment duration, gonadotrophin consumption and ovarian response is still a matter of debate. It is not clear whether an absolute LH deficiency exists following GnRH agonist downregulation. In most cases, an absolute LH deficiency really

5 Figure 2. Mean ± SE of serum hormone concentrations in both study groups during gonadotrophin treatment. There were no patients on treatment beyond day 16 in the HMG group. does not exist, as demonstrated by a very different steroidogenic response to FSH alone (Loumaye et al., 1997; Balasch et al., 2001). Notwithstanding this, a need for some LH supplementation may be evidenced in some women, depending on the extent to which the endogenous serum LH is suppressed by concomitant GnRH agonist therapy (Lisi et al., 2001; Westergaard et al., 2001; Tesarik and Mendoza, 2002). In fact, most of the present controversy in the literature about the value of LH levels in controlled ovarian hyperstimulation for assisted reproduction could be explained on the basis of the type and formulation of the GnRH agonists used (Westergaard et al., 2001; Balasch, 2002; Filicori, 2002). This is well supported by the studies by Janssens et al. (1998, 2000). Those authors first reported a clear dose-dependency of pituitary desensitization with GnRH agonists during controlled ovarian hyperstimulation in IVF and concluded that in a so-called long protocol it is possible to adjust the dose of agonist to lower values than routinely given, while adequate desensitization of the pituitary at the time of HCG injection is maintained (Janssens et al., 1998). Later, the same group of researchers performed the one and only prospective, randomized, doubleblind, placebo-controlled, dose-finding study for the GnRH agonist triptorelin in IVF (Janssens et al., 2000). Their results showed that the currently used GnRH agonists dosages in assisted reproduction treatment are far too high, resulting in unphysiological LH concentrations. According to that study, a reduction of the usual daily dose of triptorelin by 50%, and even 85%, is possible without negatively affecting the success rate (Janssens et al., 2000). Since long-acting GnRH agonist preparations produce profound, sustained desensitization (Yim et al., 2001), a single injection of 3.75 mg depot triptorelin was used in a recent study (Filicori et al., 2001) and in the present investigation to compare the ovarian responses to FSH-only products or HMG in normo-ovulatory GnRH agonist suppressed women. Also, in both studies, a fixed menotropin dose was used for at least 14 consecutive days to provide a constant FSH input to all patients and thus isolate the impact of LH activity supplementation from potential FSH dose-related effects. However, it should be noted that there was an overall 17% cancellation rate among the 60 patients initially recruited in the present study, which seems too high for a relatively young IVF population (as many as 46, or 77%, of the 60 women were younger than 36 years). Therefore, the two ampoules per day gonadotrophin starting dose in patients undergoing pituitary desensitization with a depot GnRH agonist for assisted reproduction used for the specific purpose of this study, does not seem a recommendable approach in a general IVF programme. Using this experimental, rather than a clinically useful, treatment regimen it was found, as in the previous study (Filicori et al., 2001), that HMG administration is associated with a significant reduction of treatment duration, together with increased concentrations of oestradiol in serum. This may be explained by increased LH stimulation of theca-granulosa cell function in the HMG group. The following facts support this contention. First, as expected, the AUC of serum LH was 39

6 Table 2. Gonadotrophin treatment, ovarian response and intracytoplasmic sperm injection (ICSI) outcome in groups rhfsh and HMG. Values are means ± SE unless otherwise stated. Variable rhfsh group HMG group P-value (n = 25) (n = 25) Days of stimulation ± ± 0.44 <0.05 Ampoules of rhfsh/hmg ± ± 1.01 <0.05 No. of follicles on HCG day < 10 mm 3.53 ± ± 0.69 NS mm 4.95 ± ± 0.68 NS >14 17 mm 9.42 ± ± 0.99 NS >17 mm 6.89 ± ± 0.47 <0.02 Oestradiol on HCG day (pg/ml) 1981 ± ± 236 NS Endometrial thickness on HCG day (mm) 10.9 ± ± 0.9 NS Patients with HCG and oocyte retrieval (n) 25 (100) 25 (100) NS No. oocytes retrieved ± ± 0.87 <0.05 No. metaphase II oocytes ± ± 0.84 <0.02 No. 2PN oocytes on day ± ± 0.69 <0.01 Fertilization rate on day 1 (%) <0.01 No. cleaved embryos on day ± ± 0.55 <0.01 Multinucleated embryos 0.69 ± ± 0.39 NS No. high quality embryos 6.06 ± ± 0.44 <0.02 No. embryos cryopreserved 1.95 ± ± 0.51 NS No. embryos per replacement 2.42 ± ± 0.15 NS High quality embryos replaced (%) <0.01 No. patients with embryo transfer a 25 (100) 25 (100) NS Implantation rate (%) NS Clinical pregnancies Number 9 9 NS Per oocyte retrieval (%) NS Per embryo transfer (%) NS Miscarriages 1 (11.1) 3 (33.3) NS Values in parentheses are percentages. rhfsh = recombinant human FSH; HMG = human menopausal gonadotrophin; HCG = human chorionic gonadotrophin; 2PN = two pronuclear; NS = not significant. a Values are relative to the number of patients with oocyte retrieval. Table 3. Serum hormone concentrations during the first 16 days of ovarian stimulation in groups rhfsh and HMG, calculated in each cycle as the area under the curve (AUC). Values are means ± SE. Hormone rhfsh group HMG group P-value (n = 25) (n = 25) FSH ± ± 9.2 NS LH 24.1 ± ± 3.1 <0.05 Oestradiol 4511 ± ± 513 <0.01 Androstenedione 1966 ± ± Testosterone ± ± 84 NS Progesterone 13.6 ± ± 1.54 NS Inhibin A ± ± 94 NS Inhibin B 6937 ± ± 651 <0.01 Inhibin B:inhibin A ratio ± ± 45 NS rhfsh = recombinant human FSH; HMG = human menopausal gonadotrophin; NS = not significant. 40

7 higher in group HMG. Second, paracrine signalling activated by FSH and LH sustains growth and oestrogen secretion by the developing follicle until ovulation (Hillier, 2001). Third, increased concentrations of serum androstenedione were found in the HMG group; androstenedione, a direct secretory product of LH action on thecal cells, was not measured in the study by Filicori et al. (2001). Finally, the AUC for inhibin B was increased in the HMG group versus the rhfsh group in the face of similar patterns of folliculogenesis (as shown by serial ultrasonographic scans and the inhibin B:inhibin A ratio during gonadotrophin treatment). A recent study showed that the combination of oestradiol (significantly increased in group HMG in the present study) and rhfsh resulted in a greater increase in inhibin B from small antral follicles than did rhfsh administration alone (Welt et al., 2001). However, a major influence on granulosa cell function, including inhibin formation, is exerted by androgens produced by LH-stimulated theca cells (Hillier, 2001). Androgens thus serve both as oestrogen precursors and paracrine regulators of granulosa cell function. Therefore, locally produced steroidal and nonsteroidal factors that are under LH control promote the FSH responsiveness of immature follicles (Hillier, 2001). As expected, HCG could not be detected in the serum of patients receiving rhfsh, but by contrast with the study by Filicori et al. (2001), HCG values were also below the detection limit of the assay throughout the present study period in patients treated with HMG. The higher sensitivity of the assay for β-hcg used in the previous study (0.1 IU/l versus 2 IU/l of the assay used in the present study) (Filicori et al., 2001) may account for the lack of consistency with results in the present investigation. This could also be related in part to the lack of batch to batch consistency of urinary preparations and to the fact that some LH activity is replaced by HCG in some batches of HMG preparations, including Menogon used in the study by Filicori et al. (2001) and Pergonal used in the present study. For instance, Menopur (considered a purified HMG preparation) has been reported to be unique in deriving the majority of its activity from HCG rather than LH (Giudice et al., 2001). As previously discussed (Balasch and Barri, 2001), the number of oocytes obtained would be the best efficacy parameter to evaluate when investigating a gonadotrophin in IVF, since it is the primary objective and the most direct result of ovarian stimulation with gonadotrophins and the parameter most easily observed and assessed. Using a step-down regimen of gonadotrophin administration in two large comparative studies comparing pure or highly purified FSH versus HMG in an IVF programme, the present authors reported that HMG use does not improve folliculogenesis or oocyte yield (Balasch et al., 1996). However, using the protracted stimulation regimen described above, the present study showed that rhfsh is more efficacious than HMG in terms of number of leading follicles on the day of HCG injection and oocytes obtained, even in IVF patients down-regulated with a depot GnRH agonist preparation. Subtle differences at the level of the oligosaccharide moieties of the molecules and differences in isohormone composition has been proposed to explain the higher in-vivo biological activity of recombinant gonadotrophins compared with that of urinary products (Lambert et al., 1998; Hugues, 2001; Hugues et al., 2001; Katzorke et al., 2001; Risquez, 2001; Andersen, 2002; Gordon, 2002; Nayudu et al., 2002). Less acidic species in the recombinant preparations would be more potent than the more acidic glycoforms found in the very acidic urinary gonadotrophins in terms of a more effective activation of the hormone receptor (Lambert et al., 1998). The higher biopotency of the recombinant gonadotrophin in the present study is also supported by the increased number of mature oocytes and good quality zygotes and embryos obtained in the rhfsh group compared with that of patients receiving HMG. In this respect, it is worthy to note that while analysing the effects of LH suppression during ovarian stimulation for IVF, Fleming et al. (2000) reported that with the more potent rhfsh treatment, suppression of LH failed to show the impact upon ovarian responses that were evident with highly purified urinary FSH; in fact, the same rhfsh preparation was used in Fleming s study and the present one. Finally, the finding that there were fewer large follicles on the day of HCG injection in patients receiving HMG could be related to the intriguing hypothesis postulated about hypogonadotrophic hypogonadal women treated with rfsh and rlh (The European Recombinant Human LH Study Group, 1998) and suggest that an LH ceiling effect may exist, i.e. some secondary follicles undergo atresia due to their high sensitivity to LH activity (Hillier, 1994). However, it is not known which LH dose included in the HMG preparation is sufficient to explain such an effect (Hillier, 2000). In conclusion, the present study has shown that in conditions of profound LH suppression, supplemental LH may be required in terms of treatment duration and gonadotrophin consumption. Thus, using such a protocol, especially in comparative trials, can lead to biases in favour of LHcontaining gonadotrophin preparations. In spite of this, both oocyte and embryo yield and quality were significantly higher with the use of rhfsh compared with HMG. However, while similar to that reported in the previous study (Filicori et al., 2001), the relatively small number of patients included in the present investigation precludes any conclusion on the consequences on pregnancy rate and outcome. Therefore, further large multicentre studies are warranted. References Andersen CY 2002 Effect of FSH and its different isoforms on maturation of oocytes from pre-ovulatory follicles. Reproductive BioMedicine Online 5, Balasch J 2002 The potential value of mid-follicular LH. Human Reproduction 17, Balasch J, Barri PN 2001 Reflections on the cost-effectiveness of recombinant FSH in assisted reproduction. The clinician s perspective. Journal of Assisted Reproduction and Genetics 18, Balasch J, Fábregues F, Creus M et al Pure and highly purified follicle-stimulating hormone alone or in combination with human menopausal gonadotrophin for ovarian stimulation after pituitary suppression in in-vitro fertilization. Human Reproduction 11, Balasch J, Fábregues F, Peñarrubia J et al Follicular development and hormonal levels following highly purified or recombinant follicle-stimulating hormone administration in ovulatory women and WHO group II anovulatory infertile patients. Journal of Assisted Reproduction and Genetics 15, Balasch J, Vidal E, Peñarrubia J et al Suppression of LH during ovarian stimulation: analysing threshold values and effects on ovarian response and the outcome of assisted reproduction in 41

8 42 down-regulated women stimulated with recombinant FSH. Human Reproduction 16, Barbieri RL, Hornstein MD 1999 Assisted reproduction-in vitro fertilization success is improved by ovarian stimulation with exogenous gonadotropins and pitutitary suppression with gonadotropin-releasing hormone analogues. Endocrine Reviews 20, Cohen J 2003 A short review of ovarian stimulation in assisted reproductive techniques. Reproductive BioMedicine Online 6, Creus M, Ordi J, Fábregues F et al Mid-luteal serum inhibin-a concentration as a marker of endometrial differentiation. Human Reproduction 16, European Recombinant Human LH Study Group 1998 Recombinant human luteinizing hormone (LH) to support recombinant human follicle-stimulating hormone (FSH)-induced follicular development in LH- and FSH-deficient anovulatory women: a dose finding study. Journal of Clinical Endocrinology and Metabolism 83, Filicori M 2002 The potential value of mid-follicular LH. Human Reproduction 17, Filicori M, Cognigni GE, Taraborrelli S et al Luteinizing hormone activity in menotropins optimizes folliculogenesis and treatment in controlled ovarian stimulation. Journal of Clinical Endocrinology and Metabolism 86, Fleming R, Rehka P, Deshpande N et al Suppression of LH during ovarian stimulation: effects differ in cycles stimulated with purified urinary FSH and recombinant FSH. Human Reproduction 15, Giudice E, Crisci C, Altarocca V, O Brien M 2001 Characterisation of a partially purified human menopausal gonadotropin preparation. Journal of Clinical Research 4, Gordon K 2002 New developments in gonadotrophin pharmacology. Reproductive BioMedicine Online 5, Hillier SG 1994 Current concepts of the roles of follicle stimulating hormone and luteinizing hormone in folliculogenesis. Human Reproduction 9, Hillier SG 2000 Controlled ovarian stimulation in women. Journal of Reproduction and Fertility 120, Hillier SG 2001 Gonadotropic control of ovarian follicular growth and development. Molecular and Cellular Endocrinology 179, Hugues JN 2001 Recombinant human follicle-stimulating hormone: a scientific step to clinical improvement. Reproductive BioMedicine Online 2, Hugues JN, Bständig B, Bry-Gauillard H et al Comparison of the effectiveness of recombinant and urinary FSH preparations in the achievement of follicular selection in chronic anovulation. Reproductive BioMedicine Online 3, Jacobson A, Marshall JR 1969 Ovulatory response rate with human menopausal gonadotropins of varying FSH-LH ratios. Fertility and Sterility 20, Janssens RMJ, Vermeiden JPW, Lambalk CB et al Gonadotrophin-releasing hormone agonist dose-dependency of pituitary desensitisation during controlled ovarian hyperstimulation in IVF. Human Reproduction 13, Janssens RMJ, Lambalk CB, Vermeiden JPW et al Dosefinding study of triptorelin acetate for prevention of a premature LH surge in IVF: a prospective, randomized, double-blind, placebo-controlled study. Human Reproduction 15, Katzorke T, Verhoeven HC, Blechschmidt J et al Follitropinalfa for ovarian stimulation during assisted reproduction treatment: a national collaborative study. Reproductive BioMedicine Online 3, Lambert A, Talbot JA, Anobile CJ, Robertson WR 1998 Gonadotrophin heterogeneity and biopotency: implications for assisted reproduction. Molecular Human Reproduction 4, Lisi F, Rinaldi L, Fishel S et al Use of recombinant FSH and recombinant LH in multiple follicular stimulation for IVF: a preliminary study. Reproductive BioMedicine Online 3, Loumaye E, Engrand P, Howles CM, O Dea L 1997 Assessment of the role of serum luteinizing hormone and estradiol response to follicle-stimulating hormone on in vitro fertilization outcome. Fertility and Sterility 67, Louwerens B 1969 The clinical significance of the FSH/LH ratio in gonadotropin preparations of human origin. A review. Acta Obstetricia et Gynecologica Scandinavica 48 (suppl. 1), Ludwig M, Felberbaum RE, Diedrich K, Lunenfeld B 2002 Ovarian stimulation: from basic science to clinical application. Reproductive BioMedicine Online 5 (suppl. 1), Mendoza C, Cremades N, Ruiz-Requena E et al Relationship between fertilization results after intracytoplasmic sperm injection, and intrafollicular steroid, pituitary hormone and cytokine concentrations. Human Reproduction 14, Nayudu PL, Vitt UA, Barrios de Tomasi J et al Intact follicle culture: what it can tell us about the roles of FSH glycoforms during follicle development. Reproductive BioMedicine Online 5, Peñarrubia J, Balasch J, Fábregues F et al Day 5 inhibin B serum concentrations as predictors of assisted reproductive technology outcome in cycles stimulated with gonadotrophinreleasing hormone agonist-gonadotrophin treatment. Human Reproduction 15, Risquez F 2001 Induction of follicular growth and ovulation with urinary and recombinant gonadotrophins. Reproductive BioMedicine Online 3, Tesarik J, Mendoza C 2002 Effects of exogenous LH administration during ovarian stimulation of pituitary down-regulated young oocyte donors on oocyte yield and developmental competence. Human Reproduction 17, Veeck LL 1999 An Atlas of Human Gametes and Conceptuses. Parthenon, New York, USA. Welt CK, Smith ZA, Pauler DK, Hall JE 2001 Differential regulation of inhibin A and inhibin B by luteinizing hormone, folliclestimulating hormone, and stage of follicle development. Journal of Clinical Endocrinology and Metabolism 86, Westergaard LG, Erb K, Laursen SB et al Human menopausal gonadotropin versus recombinant follicle-stimulating hormone in normogonadotropic women down-regulated with a gonadotropinreleasing hormone agonist who were undergoing in vitro fertilization and intracytoplasmic sperm injection: a prospective randomized study. Fertility and Sterility 76, Yim SF, Lok IH, Cheung LP et al Dose-finding study for the use of long-acting gonadotrophin-releasing hormone analogues prior to ovarian stimulation for IVF. Human Reproduction 16, Received 5 March 2003; refereed 24 March 2003; accepted 28 March 2003.