In-vivo ovarian androgen responses to recombinant FSH with and without recombinant LH in polycystic ovarian syndrome

Anthony P. Cheung1,2,3, Sheila M. Pride2, Basil Ho Yuen2 and Lydia Sy2

1 Department of Obstetrics and Gynaecology, University of Alberta, Edmonton and 2 Department of Obstetrics and Gynaecology, University of British Columbia, Vancouver, Canada


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
BACKGROUND: Effects of exogenous LH on ovarian androgen secretion during ovulation induction have not been clearly characterized in polycystic ovarian syndrome (PCOS). The purpose of this study was to compare androgen secretion in PCOS women during ovarian stimulation with either recombinant FSH (rFSH) alone or combined with recombinant LH (rLH). METHODS: Clomiphene-resistant women with PCOS were allocated, in a factorial study design, to receive either daily injections of rFSH (n = 24) or rFSH + rLH (n = 24) in a 1:1 ratio starting: (i) on day 2–3 of progestogen-induced menses (n = 8); (ii) after 6 weeks of GnRH agonist treatment (nafarelin, 400 µg twice daily; n = 8); or (iii) after nafarelin treatment as in (ii) plus dexamethasone (n = 8). The effects of rFSH with rFSH + rLH under these three hormone conditions on serum LH, 17{alpha}-hydroxyprogesterone (17-OHP), androstenedione ({Delta}{Delta}4) and testosterone were contrasted by analysis of variance with specific treatment days as a repeated measures factor. RESULTS: Pre-study hormone levels were similar for all groupings. Nafarelin significantly suppressed LH levels, which remained at the lower limit of assay sensitivity (0.5 IU/l) during stimulation with rFSH but increased significantly to >1 but <2 IU/l when rLH was added. As expected, 17-OHP, {Delta}{Delta}4 and testosterone levels fell following nafarelin treatment. Dexamethasone further suppressed 17-OHP, {Delta}{Delta}4 and testosterone levels and unmasked a small but significant rise in these ovarian steroids 24 h following the first dose of rFSH + rLH, a rise that was absent with rFSH alone. Secretion of these steroids then appeared to ‘catch-up’ after 5 days of rFSH stimulation. CONCLUSIONS: Despite profound LH, 17-OHP, {Delta}{Delta}4 and testosterone suppression, comparable E2 response, follicle development and successful pregnancies in PCOS subjects receiving rFSH alone to those receiving rFSH + rLH would argue that circulating LH at levels as low as 0.5 IU/l are sufficient to sustain adequate follicle development and function when FSH is present in abundance. Whether the observed dichotomy between rFSH and rFSH + rLH treatment in temporal secretion patterns reflects a greater reliance on evolving paracrine mechanisms as the follicles mature under profound LH suppression remains to be explored but may influence the optimal LH threshold for ovulation induction in PCOS.

Key words: androgens/ovarian stimulation/PCOS/recombinant FSH/recombinant LH


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Polycystic ovarian syndrome (PCOS) is a common endocrinopathy characterized by chronic anovulation and clinical or biochemical hyperandrogenism (Lobo, 1995Go; Yen, 1999Go). Elevated circulating androgens are of predominantly ovarian origin, but an adrenal component is not uncommon (Azziz et al., 1998Go). Because androgen production by ovarian thecal cells is LH-dependent, enhanced ovarian androgen output in PCOS has been attributed to inappropriate elevation of circulating LH levels, a frequent finding (up to 70%) in women with PCOS (Fox et al., 1991Go). Ovarian androgen secretion is further augmented by hyperinsulinaemia which is another common feature of PCOS (Nestler, 1997Go). While high androgen micro-environment has been correlated with atretic follicles (Teissier et al., 2000Go), other studies suggest a trophic role for androgen in pre-antral and antral follicle growth (in addition to its well-recognized role as an obligatory substrate for estradiol synthesis) and provide some evidence for positive, complementary interactions between FSH and androgen effects on follicle development (Weil et al., 1999Go).

Clinically, persistent elevation of LH during the follicular phase has been associated with decreased pregnancy rates and increased spontaneous pregnancy loss rates in PCOS, (Homburg et al., 1988Go; Regan et al., 1990Go) although other investigators have not found any significant differences in pregnancy rates (Larsen et al., 1990Go; Sagle et al., 1991Go). Nevertheless, these associations underpin the rationale for using highly purified urinary FSH (containing negligible amounts of LH) and more recently, recombinant FSH (devoid of LH), for ovulation induction in women with PCOS instead of hMG which contain FSH:LH in a 1:1 ratio. On the other hand, profound suppression of LH has been negatively associated with poorer oocyte retrieval and fertilization indices in women undergoing IVF treatment (Fleming et al., 1998Go; Anderiesz et al., 2000Go; Westergaard et al., 2000Go). Currently, the optimal LH range during gonadotrophin treatment for the various clinical categories, including PCOS, is unknown; nor is the impact of altered LH levels on steroidogenesis, particularly ovarian thecal androgen production that may impact on normal follicle development. The recent availability of recombinant LH (rLH) and FSH (rFSH) preparations has provided clinical in-vivo probes to examine both the individual and combined effects of these two gonadotrophins on ovarian androgen secretion.

The main purpose of this study, therefore, was to compare androgen secretion in PCOS women, during ovarian stimulation with either rFSH alone or combined with rLH in a 1:1 ratio, similar to that found in hMG. Previous studies examining the additional effects of LH by comparing highly purified urinary FSH with hMG have been hampered by residual LH activity and/or measurable hCG in these two gonadotrophin preparations (Larsen et al., 1990Go; Sagle et al., 1991Go; Filicori et al., 2001Go). Because small changes in circulating LH levels following rLH administration and corresponding thecal androgen secretion could be masked by endogenous LH release and adrenal androgen production respectively, this study included subgroups of women with PCOS pre-treated with GnRH agonist (to suppress pituitary LH release) and GnRH agonist combined with dexamethasone [to suppress adrenocorticotrophic hormone (ACTH) and adrenal androgens] to specifically examine these important hormonal interactions.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Subjects
Forty-eight patients with PCOS, who failed to ovulate with clomiphene citrate up to 250 mg daily were recruited for this study. We defined PCOS according to the widely accepted North American NIH 1990 consensus of hyperandrogenism of ovarian origin and oligomenorrhoea or amenorrhoea with exclusion of other specific disorders (Lobo, 1995Go; Zawadzki and Dunaif, 1995Go). All our subjects had elevated serum testosterone, while some had elevated androstenedione ({Delta}4) as well. All had a baseline, pre-study transvaginal ultrasound showing findings typical of polycystic ovaries (Adams et al., 1986Go). Pre-treatment demographic and endocrine characteristics were similar with no differences among the various groups defined below (Table IGo).


View this table:
[in this window]
[in a new window]
 
Table I. Pre-study clinical and laboratory data
 
Study drugs
Recombinant human LH and FSH (Luveris® and Gonal-F® containing 75 IU/ampoule of rLH and rFSH respectively) and human chorionic gonadotrophin (hCG; Profasi®) were kindly donated by Serono Canada Inc. The GnRH agonist, nafarelin (Synarel®) was kindly provided by Searle Canada. Dexamethasone (Dexasone®) was given by ICN Canada Ltd. Recombinant FSH alone or in combination with rLH was administered subcutaneously. Each prescribed gonadotrophin dose was reconstituted in 1 ml of water for injection into the anterior abdominal wall, below the umbilicus. Subjects on both rFSH and rLH, therefore, received two separate injections in succession at nearby sites.

Study design and treatment protocol
This was a prospective, controlled study conducted at a university-based fertility centre. The study was approved by the Research Ethics Committee of the University of Alberta and informed consent was obtained from all subjects. Forty-eight subjects were recruited for this study. The study design was a 2x3 (between-group) balanced factorial design (Winer, 1971Go) with time as a repeated-measures component (Figure 1Go). The first between-group factor of the factorial design was the two gonadotrophin regimens, rFSH alone (n = 24) versus rFSH + rLH in a 1:1 ratio (n = 24) as in conventional hMG preparations. These two gonadotrophin regimens were then further divided into three subgroups (the second between-group factor) as follows: (i) gonadotrophin stimulation alone (‘None’); (ii) nafarelin suppression (‘Naf’); and (iii) nafarelin suppression with dexamethasone (‘Naf + Dex’).



View larger version (12K):
[in this window]
[in a new window]
 
Figure 1. Schematic diagram of the 2x3 (between-group) balanced factorial study design.

 
The time-points of interest (the repeated measures factor) are defined as follows: (i) ‘day 0', the first day of gonadotrophin treatment immediately before any gonadotrophin injections; (ii) ‘day 1’, 24 h after the first dose of rFSH or rFSH + rLH; (iii) ‘day 5', 24 h after the fifth dose of rFSH or rFSH + rLH in the fixed-dose phase (see below); (iv) ‘day hCG – 1’, the day preceding hCG adminstration; and (v) ‘day hCG’, the day of hCG administration. The main parameters of interest were circulating levels of LH, 17{alpha}-hydroxyprogesterone (17-OHP), androstenedione ({Delta}4) and testosterone at the above time-points. By simultaneously comparing the two main gonadotrophin regimens under three different hormone backgrounds at the time-points of interest, this factorial experimental design had the distinct advantage of being able to study the dynamic interactions between these factors (the main focus of interest in this study) with economy of sample size, information that would otherwise be unavailable if each subgroup experiment was considered a separate entity (Winer, 1971Go).

All subjects had a transvaginal ultrasound evaluation in addition to a serum progesterone test to confirm anovulation before initiating any hormone treatment. Subjects in the ‘None’ subgroup began gonadotrophin stimulation on the second or third day of withdrawal bleeding following a 10 day course of medroxyprogesterone acetate 10 mg per os daily (Provera®; Upjohn). In the ‘Naf’ and ‘Naf + Dex’ subgroups, intranasal nafarelin was started at the higher dose of 400µg twice a day for 6 weeks, to ensure maximal pituitary–ovarian suppression before gonadotrophin administration. On the first day of gonadotrophin therapy, nafarelin was reduced to the standard dose of 200 µg twice daily and continued up to the day of hCG. Dexamethasone 0.5 mg per os four times daily was initiated 4 days before gonadotrophin stimulation; on the second day of stimulation the dosage was reduced to 0.5 mg nightly and continued until the onset of menstrual flow or confirmation of a pregnancy. There were two phases of gonadotrophin induction. During the initial 5 days of stimulation, subjects received a fixed dose of 150 IU rFSH (2 ampoules) daily with or without the equivalent dose of rLH. In the next phase, the gonadotrophin dosage was adjusted according to serial plasma estradiol (E2) and ultrasound monitoring. Doses were adjusted by 0.5–1 ampoules (37.5–75 IU) when the exponential rise in E2 was anticipated, based on the trend of E2 and ultrasound results from preceding days. Twenty-four hours after the last gonadotrophin dose, the patient received hCG 10 000 IU when there was one leading follicle with a minimal mean diameter of 15 mm.

Blood samples were drawn at 0800 h following 20 min of rest at the Clinical Investigation Unit of the University of Alberta. Just before receiving the first gonadotrophin dose, three baseline samples were collected 20 min apart between 0700 and 0800 hours and were pooled for subsequent hormone assays. All samples were taken after an overnight fast and before the morning dose of nafarelin and gonadotrophins. An aliquot of each subject’s serum sample was immediately used for E2 measurement in accordance with the clinical protocol outlined above; the remaining serum was stored at –20°C until gonadotrophin and androgen assays at study completion.

Hormone assays
LH and FSH levels were measured by microparticle enzyme immunoassay based on chromofocusing techniques (ImxTM; Abbott Diagnostics, IL, USA) (Cheung and Chang, 1995Go) which incorporates two high-affinity antibodies for higher sensitivity and specificity compared to the traditional radioimmunoassay. The assay sensitivity for LH was 0.5 IU/l and the intra- and inter-assay coefficients of variation (CV) were 3–6 and 6% respectively. Corresponding values for FSH were 0.2 IU/l, 2 and 4–6%. Steroids were measured using commercial radioimmunoassay kits: 17-OHP, testosterone and E2 from Diagnostic Products Corporation (Los Angeles, CA, USA) and {Delta}4 from Diagnostic Systems Laboratories, Inc. (Webster, TX, USA) which had assay sensitivities of 0.21 nmol/l, 0.13 nmol/l, 29 pmol/l and 0.10 nmol/l respectively. The corresponding intra- and inter-assay CV were 3.5–6.7 and 5–11% for 17-OHP; 3.8–11 and 6.4–11% for testosterone; 4.0–7.0 and 4.2–8.1% for E2; and 2.4–6.3 and 8.2–12.5% for {Delta}4.

Statistical analysis
The results were reported as means ± SEM. Between-gonadotrophin regimens and between-subgroup differences and their interactions were compared by analysis of variance (ANOVA). Time-related differences in hormone levels were evaluated by repeated measures ANOVA. Significant overall effects were followed by pairwise comparison using the Bonferroni t-procedure. Specific cycle parameters were also compared by ANOVA: the duration of gonadotrophin induction (days); the total number of ampoules of rFSH required; and on the day of hCG injection, the number of mature (>=15 mm) follicles, medium-sized (10–14 mm) follicles and endometrial thickness on ultrasound assessment and the E2 levels. Statistical analysis was performed using Statistica® 6.0 (Statsoft Inc., 2001; Tulsa, OK, USA) and the level of significance was defined at P < 0.05 (two-tailed).


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Pre-study and baseline hormone levels
Pre-study (i.e. before medroxyprogesterone or nafarelin treatment) clinical and laboratory data were not significantly different among subjects allocated to the two gonadotrophin regimens of ‘rFSH’ or ‘rFSH + rLH’ (Table IGo) or among the three subgroups of ‘None’, ‘Naf’ and ‘Naf + Dex’ (results not shown). Baseline hormones immediately before administering the first dose of gonadotrophins also were not different between the two gonadotrophin regimens overall but were significantly different among the three subgroups of ‘None’, ‘Naf’ and ‘Naf + Dex’ (Table IIGo). As expected, LH levels were significantly lower following nafarelin suppression with no further differences when dexamethasone was added (Table IIGo). Levels of 17-OHP, {Delta}4 and testosterone were also significantly lower after nafarelin suppression, but were further suppressed when dexamethasone was added (Table IIGo). Note that baseline LH levels in ‘None’ subgroups (Table IIGo) were lower than pre-study values (Table IGo), which were attributed to the suppressive effects of medroxyprogesterone acetate. Baseline FSH or E2 levels were also comparable in both gonadotrophin regimens but were significantly lower following nafarelin suppression and were not affected by the addition of dexamethasone (Table IIGo).


View this table:
[in this window]
[in a new window]
 
Table II. Baseline hormone parameters immediately before gonadotrophin treatment
 
Hormone levels following gonadotrophin stimulation
Irrespective of the gonadotrophin regimens or the three subgroups (‘None’, ‘Naf’ and ‘Naf + Dex’), there were significant increases observed in levels of LH, 17-OHP, {Delta}4 and testosterone during ovarian stimulation (P < 0.0001 for all hormones). Analysis of the complete factorial model showed that LH levels were significantly higher following rFSH + rLH than rFSH treatment (P < 0.01); however, these differences were significantly influenced by whether or not subjects received nafarelin (P < 0.0001). Similarly, differences in 17-OHP, {Delta}4 and testosterone levels at the different time-points between rFSH and rFSH + rLH treatment were significantly dependent on whether or not subjects received nafarelin or nafarelin with dexamethasone (P values were all < 0.001 for 17-OHP, {Delta}4 and testosterone). Thus, a dichotomous pattern according to rFSH and rFSH + rLH treatment was observed only in the two subgroups which received nafarelin (‘Naf’ and ‘Naf + Dex’ in Figures 2 and 3GoGo).



View larger version (19K):
[in this window]
[in a new window]
 
Figure 2. Time-course for LH levels following stimulation with rFSH or rFSH + rLH (in a 1:1 ratio) according to whether subjects received nafarelin (Naf), nafarelin with dexamethasone (Naf + Dex) or neither (None). The time-course shows hormone levelsat baseline (day 0), 24 h after the first gonadotrophin injection (day 1), 24 h after five fixed doses of gonadotrophin injections (day 5), on the day before hCG (day hCG – 1), and on the day of hCG (day hCG). Values are the means ± SEM. Significant differences according to ANOVA in the overall factorial model were followed by the Bonferonni t-procedure for pairwise comparison. Significant differences between rFSH and rFSH + rLH at each time-point are indicated (*) at the top of each graph panel. Significant time-related differences for each regimen relative to day 0, day 1, day 5, day hCG – 1 are highlighted by superscripts ‘a’, ‘b’, ‘c’ and ‘d’ respectively. Note differences in the y-axis scales between ‘None’ and the other two subgroups.

 


View larger version (33K):
[in this window]
[in a new window]
 
Figure 3. Time-course for 17{alpha}-hydroxyprogesterone, androstenedione and testosterone levels following stimulation with rFSH or rFSH + rLH (in a 1:1 ratio) according to whether subjects received nafarelin (Naf), nafarelin with dexamethasone (Naf + Dex) or neither (None). For annotations, see Figure 2Go. Again, note differences in the y-axis scales between ‘None’ and the other two subgroups.

 
Specifically, without nafarelin suppression (‘None’, Figure 2AGo), no significant differences in LH levels were detected between rFSH and rFSH + rLH treatment due to the confounding effects of endogenous LH release and the erratic premature LH surge observed in some subjects. Correspondingly, no significant differences in 17-OHP, {Delta}4 and testosterone levels between the two gonadotrophin regimens in the ‘None’ subgroup were observed (Figure 3A, D, GGo). It is interesting to note that the slightly higher LH levels in the rFSH group, although not statistically significant, were associated with higher mean values of 17-OHP, {Delta}4 and testosterone at some of the time-points. When nafarelin suppression removed this ‘noisy’ background of endogenous LH output (Figure 2B and CGo), LH levels following rFSH + rLH were significantly greater than baseline. These LH levels were also significantly greater than levels in subjects receiving rFSH alone, which remained essentially unchanged during the observation period.

Despite this measurable LH increase with rFSH + rLH stimulation in the ‘Naf’ subgroup, 17-OHP, {Delta}4 and testosterone levels were not significantly different between the two gonadotrophin regimens in the early, fixed-dose phase of the study (day 1 and day 5, Figure 3B, E, HGo). In contrast, when the adrenal contribution of these steroids was suppressed by concomitant dexamethasone treatment, there was an early rise of 17-OHP, {Delta}4 and testosterone levels on day 1 with rFSH + rLH which was absent with rFSH stimulation (‘Naf + Dex’; Figure 3C, F, IGo). There was also a trend toward higher steroid levels in the pre-ovulatory, dose-adjusted phase (day hCG – 1, day hCG) with rFSH + rLH which did not reach statistical significance, except for testosterone levels in the ‘Naf’ subgroup (Figure 3HGo) and 17-OHP levels in the ‘Naf + Dex’ subgroup (Figure 3CGo). In the latter subgroup, 17-OHP levels fell markedly on day hCG in those receiving only rFSH (Figure 3CGo).

To ensure that interpretation of the above findings was not confounded by differences in the degree of ovarian stimulation between rFSH and rFSH + rLH treatment, FSH and E2 levels were compared as a reference. As expected, significant, time-related increments in FSH and E2 levels were seen during gonadotrophin stimulation (P < 0.0001 for both) but the levels achieved were not significantly different between the two gonadotrophin regimens at any time-point; nor did nafarelin (with or without dexamethasone) influence the absolute levels of either FSH (Figure 4A, B, CGo) and E2 (Figure 4D, E, FGo). However, there was a trend to lower E2 levels with rFSH stimulation in subjects receiving nafarelin but this trend was less obvious in those also receiving dexamethasone. There was also a more rapid rate of E2 increase on day 5 in the ‘Naf + Dex’ subgroups.



View larger version (33K):
[in this window]
[in a new window]
 
Figure 4. Time-course for FSH and estradiol (E2) levels following stimulation with rFSH or rFSH + rLH (in a 1:1 ratio) according to whether subjects received nafarelin (Naf), nafarelin with dexamethasone (Naf + Dex) or neither (None). For annotations, see Figure 2Go.

 
Cycle parameters and pregnancy rates
There were no significant differences between rFSH and rFSH + rLH regimens or among the three subgroups in the duration of gonadotrophin stimulation, total amount of rFSH used, or among the number of mature, medium-sized follicles, endometrial thickness or E2 levels on the day of hCG (Table IIIGo). Although following stimulation with rFSH alone, the ‘Naf + Dex’ subgroup had more medium-sized follicles (10–14 mm) and the ‘Naf’ subgroup lower E2 level on the day of hCG, differences were not significant (Table IIIGo). The overall pregnancy rate for each gonadotrophin regimen was similar (16.7%). The small sample size did not allow for meaningful comparison of pregnancy rates, but were detailed below for completeness. There were four livebirths (including one set of twins) and one ectopic pregnancy in the rFSH group: one singleton pregnancy each in the ‘None’ and ‘Naf’ subgroups; and one twin pregnancy and one ectopic pregnancy in the ‘Naf + Dex’ subgroup. There were two livebirths and two early spontaneous abortions in the rFSH + rLH group: one early abortion in the ‘None’ subgroup; one early abortion and one livebirth in the ‘Naf’ subgroup; and one livebirth in the ‘Naf + Dex’ subgroup.


View this table:
[in this window]
[in a new window]
 
Table III. Cycle parameters on day of hCG injection
 

    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
The findings of this study have clearly demonstrated an increase in circulating LH levels following the addition of rLH to the rFSH regimen. This increase was only apparent when the background ‘noise’ from endogenous LH secretion was first suppressed with nafarelin. This would explain, in part, why previous studies had failed to detect differences in circulating LH levels between PCOS subjects treated with highly purified urinary FSH and hMG (Larsen et al., 1990Go; Sagle et al., 1991Go) The small rise in LH observed in this study was sufficient to increase ovarian secretion of 17-OHP, {Delta}4 and testosterone. However, this effect could be unequivocally demonstrated only by first eliminating endogenous LH effects on ovarian secretion of these steroids with nafarelin and by concomitantly suppressing ACTH effects on corresponding adrenal secretion with dexamethasone. Previous studies assessing ovarian androgen responses have been confounded by the presence of residual LH or hCG in various urinary gonadotrophin preparations, by the effects of endogenous LH secretion due to inadequate GnRH agonist suppression initiated in the preceding luteal phase as part of IVF treatment (Martin et al., 1997Go), by the effects of high LH levels acutely stimulated by GnRH agonist administration (Barnes et al., 1989Go), and by the use of hCG as an LH surrogate (McClamrock et al., 1991Go). Such confounding factors were meticulously eliminated from our study resulting in the unique observation that increased ovarian secretion of androgens could be demonstrated under low ambient LH levels and early in the follicular phase of an ovulation induction cycle in women with PCOS.

The differential effects of rFSH alone versus rFSH + rLH treatment have directly confirmed the role of LH in stimulating 17-OHP, {Delta}4 and testosterone secretion in vivo. Previous studies have only indirectly shown this effect using the simultaneous release of supraphysiological levels of both FSH and LH with a potent GnRH agonist (Barnes et al., 1989Go) in which LH (if using an identical immunometric assay as the current study) can increase to >100 and >40 IU/l at 4 and 24 h respectively. In contrast, LH levels following rLH administration in our study are well within the range observed in the normal menstrual cycle, and, in those subgroups pre-treated with nafarelin, are actually below the nadir observed in the luteal/early follicular phase of spontaneous cycles. LH levels in the rFSH group treated with nafarelin remained close to the assay sensitivity of 0.5 IU/l of a newer monoclonal immunometric assay based on chromofocusing techniques. When rLH was added, mean LH levels increased to between 1.37 and 1.73 IU/l (Figure 2Go). It is likely that peak LH levels were higher given the short half-life of circulating LH (Catt and Dufau, 1991Go; le Cotonnec et al., 1998Go) and that blood samples were collected 24 h after rLH injection. Combined with observations in a previous study that showed LH levels could increase by 34%, peaking at 4 h after a single i.m. injection of hMG (150 IU) (Anderson et al., 1989Go), the estimated peak levels following rLH administration in our study would still fall to <2–3 IU/l. Filicori et al. observed a similar rise in LH levels which were measured at 16 h after hMG injection in ovulatory women undergoing ovarian stimulation and intrauterine insemination (Filicori et al., 2001Go). In addition to using different study populations, their findings must be distinguished from our data by the measurable hCG in the hMG preparation used which would confound any assessment of a pure LH effect, by the inability to exclude residual LH activity in highly purified urinary FSH preparations, and finally by the lesser degree of LH suppression where baseline LH levels were >1 IU/l.

With LH and concomitant adrenal androgen suppression, we observed a difference in temporal patterns in steroid secretion. The early day 1 rise in 17-OHP, {Delta}4 and testosterone was absent when only rFSH was given (Figure 3C, F, IGo), but by day 5 these steroid levels were comparable to those observed with rFSH + rLH. In addition, despite profoundly suppressed LH levels in ‘Naf’ and ‘Naf + Dex’ subgroups stimulated with rFSH alone, a small rise in these three steroids was still observed, albeit delayed. This rise continued to day hCG – 1/day hCG (Figure 3C, F, IGo). One possibility for this delay might be a greater reliance on evolving paracrine mechanisms (involving factors such as inhibin, insulin-like growth factor and androgens) as follicles mature under profound LH suppression (Hillier, 1994Go, 2001Go). In monkeys, androgen receptor mRNA and FSH receptor mRNA have been co-localized and positively correlated in granulosa cells of growing follicles (Weil et al., 1999Go). Further, FSH can induce acquisition of androgen receptors whereas androgen stimulates FSH receptors in granulosa cells of small follicles (Weil et al., 1999Go). It is therefore tempting to speculate that similar interactions between FSH and androgen might also be in place in humans and compensate for suppressed LH effects during follicular development. Such interactions would potentially have a greater effect in PCOS subjects, known to have exaggerated intrinsic basal and LH-stimulated thecal androgen output (Gilling-Smith et al., 1994Go).

Defining optimal FSH and LH relationships and establishing optimal LH ranges for ovarian stimulation of various clinical categories is a therapeutic consideration receiving wider attention (Anonymous, 1998Go). Hillier et al. first described the concept of an optimal LH ‘ceiling’ for the follicular phase, which, if exceeded, would result in suboptimal ovulatory function and fertility rates (Hillier, 1994Go). Westergaard et al. proposed a minimal LH requirement or threshold using a level of 0.5 IU/l to demarcate ‘profound’ and ‘normal’ LH suppression in ovulatory women undergoing IVF (Westergaard et al., 2000Go). In our study, baseline LH levels were suppressed to <=0.5 IU/l (below assay sensitivity), corresponding to Westergaard’s ‘profound’ category.

The European Recombinant Human LH Study Group has recently demonstrated the beneficial effects of adding LH to the ovarian stimulation regimen for IVF patients which underscores the clinical importance of determining optimal LH thresholds during ovarian stimulation (Anonymous, 1998Go; Anderiesz et al., 2000Go). Correspondingly, our data provide some relevant information specifically for women with PCOS. The fact that E2 response, follicle development and successful conceptions were comparable with both rFSH and rFSH + rLH regimens after pituitary suppression indicates indirectly that sufficient theca androgen secretion was present to provide substrate for adequate estrogen production and follicle growth, a finding also noted in IVF patients (Balasch et al., 2001Go). Sullivan et al. observed that once the follicle achieved a diameter of 14 mm, FSH or LH was equally effective in sustaining granulosa E2 output (Sullivan et al., 1999Go). Our findings therefore suggest that provision of continued rFSH stimulation in the face of profoundly suppressed endogenous LH levels (<=0.5 IU/l) is still adequate for follicular function and pregnancy in women with PCOS, even when ovarian and adrenal androgen secretion are concomitantly suppressed.

The lack of significantly different E2 levels and mature follicle numbers between the gonadotrophin regimens might also reflect our stimulation protocol of inducing ovulation with hCG once a follicle of >=15 mm had been identified, inter-individual variations in response to FSH stimulation relative to the subject sample size. Although not statistically significant, the increased number of medium-sized follicles in the ‘Naf + Dex’ subgroup receiving only rFSH (Table IIIGo) raises the possibility that dexamethasone may exert an anti-apoptotic effect on granulosa cells as reported in cell culture experiments (Sasson et al., 2001Go).

The abrupt increase in 17-OHP and {Delta}4 24 h after rLH was added further confirms that the corresponding enzyme system, CPY17, is indeed LH responsive as suggested by previous in-vitro observations (Bogovich and Richards, 1982Go). Furthermore, we have demonstrated that this enzyme complex is responsive in the low LH milieu of our study and in the early follicular phase of ovulation induction. Whether this enzyme system in subjects with PCOS is more sensitive to LH is being investigated in a parallel study involving normal ovulatory women. It should be noted that this study only measured steroid intermediates of the {Delta}4 pathway and that specific contributions from the {Delta}5 pathway via 3ß-hydroxysteroid dehydrogenase/isomerase for the relevant androgens in the follicular phase are unknown.

In conclusion, by contrasting the effects of rFSH and rFSH + rLH with and without nafarelin, and nafarelin with and without dexamethasone, we have successfully demonstrated a significant rise in LH when rLH is added during ovulation induction. This LH increase, although small, is sufficient to stimulate 17-OHP, {Delta}4 and testosterone secretion early in folliculogenesis, an effect only revealed by first suppressing endogenous LH and ACTH. Despite profound LH and 17-OHP, {Delta}4 and testosterone suppression, comparable E2 response, follicle development and successful pregnancies in PCOS subjects receiving rFSH alone to those receiving rFSH + rLH, would argue that circulating LH at levels as low as 0.5IU/l are sufficient to sustain adequate follicular development and function when FSH is present in abundance. Definition of optimal LH ranges for various clinical conditions remains to be elucidated, but is clearly an evolving concept that could change our clinical approach to gonadotrophin therapy.


    Acknowledgements
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Dr Anthony P.Cheung would like to thank Lori Zuk, RN and her staff of the Clinical Investigator Unit at the University of Alberta Hospital; without their support and dedication, this study would not have been possible. He would also like to thank Serono Canada Inc. for the generous donations of Gonal-F®, Lhadi® and Profasi® and Searle Canada for the supply of Synarel®. The study was supported by a research grant to A.P.C. from the University of Alberta and Hospitals, Edmonton, Alberta, Canada and Serono Canada, Inc.


    Notes
 
3 To whom correspondence should be addressed at: Division of Reproductive Endocrinology and Infertility, Department of Obstetrics and Gynaecology, University of British Columbia, 2nd Floor, Willow Pavilion, Vancouver Hospital and Health Sciences Centre, 855 West 12th Avenue, Vancouver, B.C. Canada V5Z 1M9. E-mail: apcheung{at}interchange.ubc.ca Back


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Adams, J., Polson, D.W. and Franks, S. (1986) Prevalence of polycystic ovaries in women with anovulation and idiopathic hirsutism. Br. Med. J. (Clin. Res. Ed.), 293, 355–359.[ISI][Medline]

Anderiesz, C., Ferraretti, A., Magli, C., Fiorentino, A., Fortini, D., Gianaroli, L., Jones, G.M. and Trounson, A.O. (2000) Effect of recombinant human gonadotrophins on human, bovine and murine oocyte meiosis, fertilization and embryonic development in vitro. Hum. Reprod., 15, 1140–1148.[Abstract/Free Full Text]

Anderson, R.E., Cragun, J.M., Chang, R.J., Stanczyk, F.Z. and Lobo, R.A. (1989) A pharmacodynamic comparison of human urinary follicle-stimulating hormone and human menopausal gonadotropin in normal women and polycystic ovary syndrome. Fertil. Steril., 52, 216–2120.[ISI][Medline]

Anonymous (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. The European Recombinant Human LH Study Group. J. Clin. Endocrinol. Metab., 83, 1507–1514.[Abstract/Free Full Text]

Azziz, R., Black, V., Hines, G.A., Fox, L.M. and Boots, L.R. (1998) Adrenal androgen excess in the polycystic ovary syndrome: sensitivity and responsivity of the hypothalamic-pituitary-adrenal axis. J. Clin. Endocrinol. Metab., 83, 2317–2323.[Abstract/Free Full Text]

Balasch, J., Creus, M., Fabregues, F., Civico, S., Carmona, F., Puerto, B., Casamitjana, R. and Vanrell, J.A. (2001) The effect of exogenous luteinizing hormone (LH) on oocyte viability: evidence from a comparative study using recombinant human follicle-stimulating hormone (FSH) alone or in combination with recombinant LH for ovarian stimulation in pituitary-suppressed women undergoing assisted reproduction. J. Assist. Reprod. Genet., 18, 250–256.[ISI][Medline]

Barnes, R.B., Rosenfield, R.L., Burstein, S. and Ehrmann, D.A. (1989) Pituitary–ovarian responses to nafarelin testing in the polycystic ovary syndrome. New Engl. J. Med., 320, 559–565.[Abstract]

Bogovich, K. and Richards, J.S. (1982) Androgen biosynthesis in developing ovarian follicles: evidence that luteinizing hormone regulates thecal 17 alpha-hydroxylase and C17–20-lyase activities. Endocrinology, 111, 1201–1208.[ISI][Medline]

Catt, K. and Dufau, M.L. (1991) Gonadotropic hormones: biosynthesis, secretion, receptors, and actions. In Yen, S.S.C. and Jaffe, R.B. (eds), Reproductive Endocrinology. W.B.Saunders, Philadelphia, pp. 105–155.

Cheung, A.P. and Chang, R.J. (1995) Pituitary responsiveness to gonadotrophin-releasing hormone agonist stimulation: a dose–response comparison of luteinizing hormone/follicle-stimulating hormone secretion in women with polycystic ovary syndrome and normal women. Hum. Reprod., 10, 1054–1059.[Abstract]

Filicori, M., Cognigni, G.E., Taraborrelli, S., Spettoli, D., Ciampaglia, W., Tabarelli De Fatis, C., Pocognoli, P., Cantelli, B. and Boschi, S. (2001) Luteinzing hormone activity in menotropins optimizes folliculogenesis and treatment in controlled ovarian stimulation. J. Clin. Endocrinol. Metab., 86, 337–343.[Abstract/Free Full Text]

Fleming, R., Lloyd, F., Herbert, M., Fenwick, J., Griffiths, T. and Murdoch, A. (1998) Effects of profound suppression of luteinizing hormone during ovarian stimulation on follicular activity, oocyte and embryo function in cycles stimulated with purified follicle stimulating hormone. Hum. Reprod., 13, 1788–1792.[Abstract]

Fox, R., Corrigan, E., Thomas, P.A. and Hull, M.G. (1991) The diagnosis of polycystic ovaries in women with oligo-amenorrhoea: predictive power of endocrine tests. Clin. Endocrinol. (Oxf.), 34, 127–131.[ISI][Medline]

Gilling-Smith, C., Willis, D.S., Beard, R.W. and Franks, S. (1994) Hypersecretion of androstenedione by isolated thecal cells from polycystic ovaries. J. Clin. Endocrinol. Metab., 79, 1158–1165.[Abstract]

Hillier, S.G. (1994) Current concepts of the roles of follicle stimulating hormone and luteinizing hormone in folliculogenesis. Hum. Reprod., 9, 188–191.[Abstract]

Hillier, S.G. (2001) Gonadotropic control of ovarian follicular growth and development. Mol. Cell. Endocrinol., 179, 39–46.[ISI][Medline]

Homburg, R., Armar, N.A., Eshel, A., Adams, J. and Jacobs, H.S. (1988) Influence of serum luteinising hormone concentrations on ovulation, conception, and early pregnancy loss in polycystic ovary syndrome. Br. Med. J., 297, 1024–1026.[ISI][Medline]

Larsen, T., Larsen, J.F., Schioler, V., Bostofte, E. and Felding, C. (1990) Comparison of urinary human follicle-stimulating hormone and human menopausal gonadotropin for ovarian stimulation in polycystic ovarian syndrome. Fertil. Steril., 53, 426–431.[ISI][Medline]

le Cotonnec, J.Y., Porchet, H.C., Beltrami, V. and Munafo, A. (1998) Clinical pharmacology of recombinant human luteinizing hormone: Part II. Bioavailability of recombinant human luteinizing hormone assessed with an immunoassay and an in vitro bioassay. Fertil. Steril., 69, 195–200.[ISI][Medline]

Lobo, R.A. (1995) A disorder without identity: ‘HCA,’ ‘PCO,’ ‘PCOD,’ ‘PCOS,’ ‘SLS’. What are we to call it?! [editorial]. Fertil. Steril., 63, 1158–1160.[ISI][Medline]

Martin, K.A., Hornstein, M.D., Taylor, A.E., Hall, J.E. and Barbieri, R.L. (1997) Exogenous gonadotropin stimulation is associated with increases in serum androgen levels in in-vitro fertilization–embryo transfer cycles. Fertil. Steril., 68, 1011–1016.[ISI][Medline]

McClamrock, H.D., Bass, K.M. and Adashi, E.Y. (1991) Ovarian hyperandrogenism: the role of and sensitivity to gonadotropins. Fertil. Steril., 55, 73–79.[ISI][Medline]

Nestler, J.E. (1997) Insulin regulation of human ovarian androgens. Hum. Reprod., 12 (Suppl. 1), 53–62.[Medline]

Regan, L., Owen, E.J. and Jacobs, H.S. (1990) Hypersecretion of luteinising hormone, infertility, and miscarriage [see comments]. Lancet, 336, 1141–1144.[ISI][Medline]

Sagle, M.A., Hamilton-Fairley, D., Kiddy, D.S. and Franks, S. (1991) A comparative, randomized study of low-dose human menopausal gonadotropin and follicle-stimulating hormone in women with polycystic ovarian syndrome. Fertil. Steril., 55, 56–60.[ISI][Medline]

Sasson, R., Tajima, K. and Amsterdam, A. (2001) Glucocorticoids protect against apoptosis induced by serum deprivation, cyclic adenosine 3',5'-monophosphate and p53 activation in immortalized human granulosa cells: involvement of Bcl-2. Endocrinology, 142, 802–811.[Abstract/Free Full Text]

Sullivan, M.W., Stewart-Akers, A., Krasnow, J.S., Berga, S.L. and Zeleznik, A.J. (1999) Ovarian responses in women to recombinant follicle-stimulating hormone and luteinizing hormone (LH): a role for LH in the final stages of follicular maturation. J. Clin. Endocrinol. Metab., 84, 228–32.[Abstract/Free Full Text]

Teissier, M.P., Chable, H., Paulhac, S. and Aubard, Y. (2000) Comparison of follicle steroidogenesis from normal and polycystic ovaries in women undergoing IVF: relationship between steroid concentrations, follicle size, oocyte quality and fecundability. Hum. Reprod., 15, 2471–2477.[Abstract/Free Full Text]

Weil, S., Vendola, K., Zhou, J. and Bondy, C.A. (1999) Androgen and follicle-stimulating hormone interactions in primate ovarian follicle development. J. Clin. Endocrinol. Metab., 84, 2951–2956.[Abstract/Free Full Text]

Westergaard, L.G., Laursen, S.B. and Andersen, C.Y. (2000) Increased risk of early pregnancy loss by profound suppression of luteinizing hormone during ovarian stimulation in normogonadotrophic women undergoing assisted reproduction. Hum. Reprod., 15, 1003–1008.[Abstract/Free Full Text]

Winer, B. (1971) Statistical Principles in Experimental Design. 2nd edn. McGraw-Hill Inc., New York. pp. 309–430.

Yen, S.S.C. (1999) Polycystic ovary syndrome (hyperandrogenic chronic anovulation). In Yen, S.S.C., Jaffe, R.B. and Barbieri, R.L. (eds), Reproductive Endocrinology: Physiology, Pathophysiology, and Clinical Managment. W.B.Saunders, Philadelphia, pp. 436–478.

Zawadzki, J.K. and Dunaif, A. (1995) Diagnostic criteria for polycystic ovary syndrome: towards a rational approach. In Dunaif, A., Givens, J.R., Haseltine, F.P. and Merriam, G.R. (eds), Polycystic Ovary Syndrome, Current Issues in Endocrinology and Metabolism. Blackwell, Boston, pp. 377–384.

Submitted on February 15, 2002; resubmitted on June 20, 2002; accepted on July 7, 2002.





This Article
Abstract
FREE Full Text (PDF )
Alert me when this article is cited
Alert me if a correction is posted
Services
Email this article to a friend
Similar articles in this journal
Similar articles in ISI Web of Science
Similar articles in PubMed
Alert me to new issues of the journal
Add to My Personal Archive
Download to citation manager
Search for citing articles in:
ISI Web of Science (2)
Request Permissions
Google Scholar
Articles by Cheung, A. P.
Articles by Sy, L.
PubMed
PubMed Citation
Articles by Cheung, A. P.
Articles by Sy, L.