| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
From the Clinical Research Centers |
National Center for Infertility Research and Reproductive Endocrine Unit, Massachusetts General Hospital, Boston, Massachusetts 02114
Address all correspondence and requests for reprints to: Frances Hayes, Reproductive Endocrine Unit, Massachusetts General Hospital, Boston, Massachusetts 02114.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
LH levels decreased in a dose-dependent fashion after administration of the GnRH antagonist (P < 0.0001), with a maximum percent inhibition of 83 ± 2%. At all except the 5 µg/kg dose, mean LH levels remained significantly lower than baseline for up to 20 h post antagonist (P < 0.002). At all antagonist doses, both the degree and duration of LH suppression were similar in PCOS and normal women. The maximum percent inhibition of FSH was 39 ± 2%, which was significantly less than that of LH (P < 0.001). Testosterone (T) levels fell significantly within 4 h of antagonist administration, with maximum percent inhibition of 39 ± 3% occurring at 8 h. In the patients in whom 150 µg/kg of the antagonist was given for 3–7 days, no further suppression of either gonadotropins or T was noted.
Our conclusions were: 1) The equivalent susceptibility of LH to submaximal GnRH receptor blockade in normal and PCOS women suggests that the elevated LH levels in PCOS are not the result of an increase in the quantity of GnRH secreted. These data imply that it is the frequency of GnRH stimulation per se and/or enhanced pituitary sensitivity to endogenous GnRH that underlie the gonadotropin abnormalities in PCOS; and 2) The rapid suppression of T with increasing GnRH antagonist dose is consistent with acute regulation of T secretion by LH.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The relative contributions of the pituitary and hypothalamus to the
abnormal gonadotropin secretory dynamics in PCOS have been difficult to
quantify. The increase in LH pulse amplitude characteristic of PCOS (6, 7, 9, 11, 12, 13, 14) may represent either enhanced pituitary sensitivity or
increased stimulation by GnRH. Increased gonadotrope sensitivity, both
to exogenous GnRH (6, 11) and to GnRH agonists (15), indicates that the
pituitary is directly responsible for at least a part of this effect.
Most (7, 9, 12, 16), but not all (13, 17), studies have also provided
evidence for an important hypothalamic contribution by demonstrating an
increase in LH pulse frequency. Studies in which pulsatile free
-subunit (FAS) has been used as an alternate surrogate marker of
GnRH secretion have provided corroborative evidence for an increase in
pulse frequency in PCOS (18, 19). Although monitoring of pulsatile LH
and FAS secretion provides a useful insight into the frequency of GnRH
secretory episodes and has been employed extensively in the study of
GnRH physiology in the human (20, 21, 22, 23), it does not permit estimation of
the overall quantity of GnRH secreted. To address this issue, we have
previously used a selective GnRH antagonist to provide a
semiquantitative estimate of GnRH secretion in the human (24).
In the present study, the potent Nal-Glu GnRH antagonist ([Ac-D2Nal1, D4ClPhe2,-D3Pal3, Arg5, DGlu(AA)6, Dala10] GnRH) was used to assess the overall quantity of GnRH secretion in PCOS. We have previously suggested that the increased frequency of GnRH secretion in PCOS contributes to the elevated LH/FSH ratio seen in this disorder (7, 9). If this increase in frequency is accompanied by an increase in the overall amount of endogenous GnRH secreted, one would expect that LH secretion would be less susceptible to GnRH receptor blockade in women with PCOS, compared with normal women.
In PCOS, there is also evidence that the increased LH/FSH ratio contributes to the hyperandrogenism characteristic of this disorder, with the high LH levels stimulating excessive production of androgenic substrates, which in the presence of low to normal levels of FSH, are not adequately aromatized to estrogen. Though previous studies using long-acting GnRH agonists have indicated a predominantly ovarian source of androgen excess in PCOS (25, 26, 27, 28), no data is available on the time course over which androgen suppression occurs. In this study, we used complete GnRH receptor blockade to examine the impact of acute and short-term decreases in LH on gonadal steroid secretion in PCOS.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Eleven women, 24–37 yr old, with PCOS were studied (Table 1
). All had a history of amenorrhea or
oligomenorrhea, as well as clinical and/or biochemical evidence of
hyperandrogenism. Hirsutism was present in all but one subject, in whom
significant acne was a presenting feature. Though not required for
inclusion in the study, the presence of polycystic ovarian morphology
was demonstrated in all subjects by transvaginal ultrasonography, and
all had an elevated LH/FSH ratio [i.e. >2 SD
above that seen in normal women in the early follicular phase, as
previously established in our laboratory (22)]. All had a serum
progesterone (P) less than 1 ng/mL (<3.2 nmol/L) at the time they were
studied. The subjects were otherwise healthy, with normal baseline
levels of PRL, TSH and T4. Late onset congenital adrenal
hyperplasia was excluded by the presence of a normal
17-hydroxyprogesterone level in a baseline morning sample or after
stimulation with iv 0.25 mg ACTH. There was no history of excessive
exercise or use of any hormonal therapy for at least 3 months before
the study. The study was approved by the Subcommittee on Human Studies
of the Massachusetts General Hospital, and all participants provided
written informed consent.
|
Experimental protocols
To determine the acute response to GnRH receptor blockade, the patients were admitted to the General Clinical Research Center of the Massachusetts General Hospital. An iv cannula was inserted, and blood was sampled at 10-min intervals for a 4-h baseline period commencing at 0800 h. A single dose of the Nal-Glu GnRH antagonist was then administered sc at a dose of 5, 15, 50, or 150 µg/kg; after which, sampling continued at 10-min intervals for 8 h and then hourly for a further 12 h. Five to 7 PCOS patients were studied at each antagonist dose. There was no difference in the baseline hormonal profile among the groups receiving the various GnRH antagonist doses. No patient received the same dose on more than one occasion, and there was a minimum of 6 weeks between studies in any given patient. All samples were assayed for LH, whereas FSH was analyzed in hourly samples. Testosterone (T) was measured at baseline and immediately before, and every 4 h after, all doses of the GnRH antagonist. Sex hormone-binding globulin (SHBG) and dehydroepiandrosterone sulfate (DHEAS) were measured before and after the 150-µg/kg antagonist dose.
A total of 63 studies were performed in the control subjects in the early and late follicular, and early luteal phases. Because we have previously demonstrated no difference in the percent inhibition among these 3 cycle stages (24), the data from each dose during the different stages were combined, such that a total of 12, 23, 14, and 14 normal subjects were studied at the 5, 15, 50, and 150 µg/kg doses, respectively. Although gonadotropin data are presented on all controls, androgen levels are only reported for a subset of 5 women, after administration of 150 µg/kg of the antagonist, because of the limitations of sample volume.
In a second study, designed to determine the time course of gonadotropin and androgen responses to short-term GnRH receptor blockade, 150 µg/kg of the GnRH antagonist was administered sc at 24-h intervals for 3 days in seven patients and continued in a subset of three patients for a total of 7 days. Blood was drawn each day, before antagonist administration, for measurement of LH, FSH, T, DHEAS, and SHBG.
Assays
All samples from an individual study were measured in duplicate in the same assay. LH, FSH, DHEAS, and P were measured by RIA, as previously described (30, 31, 32). The intra- and interassay CVs (coefficients of variation) for LH and FSH were 6.1 and 7.0%, and 8.5 and 11.6%, respectively. The LH and FSH assays had a sensitivity of 0.8 IU/L using the Second International Reference Preparation of human menopausal gonadotropin as the standard. Serum T concentrations were determined by a previously described RIA, with a sensitivity of 12 ng/dL and an interassay CV of less than 15% (32). Androstenedione was measured, after extraction from serum, in an RIA using a polyclonal antiserum (9). The sensitivity of this assay is 0.6 ng/mL, and the intra- and interassay CVs are both less than 10%. SHBG was measured by Nichols Laboratory using an RIA.
Data analysis
To determine the significance of gonadotropin and androgen decreases after the GnRH antagonist dose, the studies were divided into a baseline 4-h period and five 4-h periods after the antagonist was administered. Mean hormone levels were then compared using ANOVA for repeated measures followed by post hoc Newman-Keuls testing for individual differences. To compare the responses of the PCOS and normal women, the data were expressed as percent baseline, and the mean hormone levels of the two groups at each antagonist dose were compared using ANOVA.
The maximum degree of gonadotropin suppression was determined by calculating the percent inhibition from the pre-antagonist period [(mean PRE - nadir)/mean PRE] x 100, as previously described (29), where nadir hormone levels were calculated using a moving average. For LH, a 6-point moving average (equivalent to 1 h of sampling) was used, because the LH nadir occurred during the 10-min sampling portion of the study. A 3-point moving average was used to calculate the FSH nadir, because the latter occurred during the hourly sampling portion of the study. The time of the nadir at each antagonist dose was compared using ANOVA. Results are expressed as mean ± SEM. P < 0.05 was taken to be statistically significant.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
LH. Acute GnRH receptor blockade was associated with a
significant reduction in mean LH levels (P < 0.002),
with a significant effect of both GnRH antagonist dose
(P < 0.01) and time (P < 0.0001; Fig. 1). The nadir for LH occurred at 5.7
± 0.7, 6.4 ± 0.6, 7.6 ± 0.5, and 9.9 ± 1.3 h
after 5, 15, 50, and 150 µg/kg, respectively, of the Nal-Glu
antagonist, with the nadir for the 150-µg/kg dose occurring
significantly later than for the two lower doses (P <
0.02). Recovery from GnRH receptor blockade was also dose dependent. At
the 5-µg/kg dose, mean LH levels began to increase approximately
8 h post antagonist, whereas at the three higher doses, mean LH
levels remained significantly lower than baseline for up to 20 h
post antagonist (P < 0.0002). At the 15-µg/kg dose,
mean LH levels at the end of the study had increased significantly from
their nadir (P < 0.005); however, complete suppression
was sustained for the entire duration of the study at the 50- and
150-µg/kg doses.
|
). This lack of difference in response
was also apparent when the data were expressed as percent inhibition
(Fig. 3). The maximum percent inhibition
of LH was 60.4 ± 1.8, 74.2 ± 1.3, 79.0 ± 4.0, and
83.1 ± 2.0%, with increasing antagonist dose (P
< 0.0001). A similar degree of suppression was achieved at the two
higher doses, which in both cases was greater than the 5-µg/kg dose
(P < 0.005); and at the 150-µg/kg dose, it was also
greater than the 15-µg/kg dose (P < 0.05). The lack
of a difference in percent inhibition at the 5-µg/kg dose
(approximately the ED50) is important to the quantification
of endogenous GnRH secretion in PCOS, relative to normal women.
|
|
). At all
doses studied, FSH levels remained significantly lower than
pretreatment for up to 20 h post antagonist (P <
0.0002). Increasing the GnRH antagonist dose had no effect on either
the time or the value of the FSH nadir, consistent with studies in
normal women (24). The time of the FSH nadir occurred significantly
later than for LH at all doses (P < 0.02). The maximum
degree of FSH suppression was significantly less than that of LH
(39 ± 2 vs. 83 ± 2% at the 150 µg/kg dose,
P < 0.001). When expressed as percent
baseline, the maximum inhibition of FSH was similar in PCOS and normal
women (Fig. 3).
Androgens. In the PCOS subjects, T levels fell in a
dose-dependent fashion after GnRH receptor blockade (P
< 0.05; Fig. 4). Maximum percent
inhibition was 38.8 ± 2.8% and occurred 8 h after the
50-µg/kg antagonist dose, by which time, T levels had fallen into the
normal range. By 20 h, mean T levels remained significantly lower
than baseline at all except the 5-µg/kg dose (P <
0.0002). There was no significant difference in the degree of T
inhibition achieved with the 50- and 150-µg/kg doses at any time
point. At baseline, there was a significant correlation between serum
LH and T (r = 0.95, P < 0.02). However, no
significant relationship was observed between percent LH inhibition and
the reduction in serum T levels. In the normal controls, the maximum
percent inhibition of T, after 150 µg/kg of the antagonist, was
similar to that seen in the PCOS subjects (28.2 ± 2.5
vs. 32.6 ± 3.2% at 8 h).
|
).
GnRH antagonist administration was not associated with a significant
change in SHBG levels in the PCOS subjects (1.1 ± 0.2
vs. 1.1 ± 0.1 mg DHT bound/dl at baseline and at
8 h post antagonist, respectively).
|
In the seven patients who received 150 µg/kg·day of the GnRH
antagonist for 3 days, mean LH values fell by 78.7 ± 3.9%
(P < 0.0001), and FSH levels by 42.1 ± 4.1%
(P < 0.002; Fig. 5),
with no difference observed in the degree of gonadotropin suppression
among the 3 days. There was a corresponding fall in T levels (21.9
± 5.6, 29.0 ± 6.6, and 27.9 ± 8.2%; P <
0.05 vs. baseline) and estradiol (27.1 ± 8.2,
21.6 ± 10.2, 32.8 ± 7.4%; P < 0.005
vs. baseline). In the three subjects in whom treatment with
the GnRH antagonist was continued for 7 days, no further suppression
was seen in either gonadotropin or androgen levels.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Direct sampling of GnRH in the hypophyseal portal blood is not feasible in the human, and measurements in the peripheral circulation do not accurately reflect hypothalamic GnRH secretion because of its rapid metabolism (34). Consequently, studies of GnRH physiology in PCOS have relied on the use of LH or FAS pulse frequency as surrogate markers of antecedent GnRH secretion. Though use of LH as a marker of GnRH pulse generator activity has been validated by careful animal studies (35, 36), this approach provides no information on the quantity of GnRH secreted. In the present study, we used a new tool to gain further insight into the neuroendocrine dynamics of PCOS by employing a GnRH antagonist to competitively block the GnRH receptor. The principles governing ligand-receptor interaction (37) predict that a semiquantitative estimate of endogenous GnRH secretion can be derived from determining the effect of competition between GnRH and a GnRH antagonist at the GnRH receptor, analogous to the previous use of naloxone to quantitate endorphin tone (38). Thus, the response of a marker of GnRH action, such as LH, can be used to assess GnRH secretion in the presence of submaximal GnRH receptor blockade, such that the amount of GnRH secreted will be inversely proportional to the degree of LH inhibition. Such an approach is only possible because GnRH is the only known secretagogue for LH, GnRH and its antagonist bind to a single receptor type, and there is no evidence of any change in GnRH receptor affinity over a wide range of both physiologic and pharmacologic conditions (24). This approach is also based on the assumption that any change in GnRH receptor number or post receptor amplification of the GnRH signal will be similar before and after acute GnRH receptor blockade and can, therefore, be accounted for by expressing the data as percent change from baseline. Although this method does not permit precise quantification of GnRH secretion, it does allow comparisons to be drawn between PCOS and normal women.
In previous studies, full dose-response curves over a range of GnRH antagonist doses were constructed in normal women (24, 29). In the present study, the same range of GnRH antagonist doses was administered to women with PCOS and was effective in suppressing LH, and to a lesser extent, FSH secretion. Both the degree and duration of LH suppression increased with increasing antagonist doses. A maximum inhibitory effect on LH was observed at the GnRH antagonist doses studied, as evidenced by the flattening of the dose-response curve at the 50- and 150-µg/kg doses, whereas the two lower doses yielded a submaximal effect. When the results were expressed as percent baseline, neither the degree nor the duration of LH suppression in PCOS differed from normal controls at any of the four antagonist doses.
The demonstration that the susceptibility of LH to GnRH receptor blockade is similar in normal and PCOS women suggests that the overall quantity of GnRH is not increased in PCOS, despite the observed increase in GnRH pulse frequency (7, 9, 12, 16) and LH pulse amplitude (6, 7, 9, 12, 13, 14). Only one other study has reported the response to a GnRH antagonist in a hyperandrogenic state. In a group of hirsute adolescent girls, the degree of LH suppression, after administration of 50 µg/kg of the Nal-Glu GnRH antagonist daily for 4 days, was similar to that in developmentally-matched controls (39). Though the results of the two studies are compatible, submaximal GnRH antagonist doses were not used in the previous study; and thus, it could not provide an assessment of GnRH secretion in PCOS, compared with normal women.
The data in the present study suggest that the abnormal neuroendocrine dynamics in PCOS cannot be explained by an increase in the overall amount of GnRH secreted. Therefore, if changes in hypothalamic GnRH secretion are involved in the gonadotropin perturbations in PCOS, it is the faster frequency of GnRH stimulation that underlies these abnormalities. We originally proposed that the elevated LH/FSH ratio in PCOS may result from the increased GnRH pulse frequency seen in this disorder, based on studies of GnRH-deficient men. In this model, an increase in the frequency of a fixed and physiologic dose of exogenous GnRH, sustained over time, resulted in a similar increase in mean LH levels with little change in FSH (40). Recent studies suggest that differential regulation of LH and FSH, by rapid frequencies of GnRH, may be explained by the observation that at fast frequencies of GnRH stimulation, there is increased pituitary production of follistatin, which blocks activin stimulation of FSH (41). The applicability of this frequency hypothesis to PCOS is supported by the correlation between GnRH-induced LH pulse frequency and both pool LH and the LH/FSH ratio (9). However, it is likely that additional factors, such as the abnormal sex steroid milieu, also contribute to the disparity in gonadotropin levels in PCOS, given that estrone administration suppresses FSH levels, while having no effect on LH (42).
Although an increase in GnRH pulse frequency may explain the elevated LH/FSH ratio in PCOS, it does not explain the increased LH pulse amplitude. Indeed, in the studies conducted in GnRH-deficient men, whereas mean LH levels increased with increasing frequency of GnRH stimulation, individual LH pulse amplitudes decreased (40). Therefore, it seems that enhanced pituitary responsiveness to GnRH is responsible for the elevated LH pulse amplitude in PCOS (6, 11). This increase in pituitary sensitivity is likely to be estrogen mediated, because estrogen is known to increase the fraction of gonadotropes that respond to GnRH (43), such that small GnRH pulses (normally insufficient to stimulate LH secretion) may result in discernible LH pulses.
In addition to probing gonadotropin dynamics, this study was also designed to examine the acute effect of LH suppression on elevated androgen levels in PCOS. Though studies employing suppressive doses of GnRH agonists (25, 26, 27, 28) point to the ovary as the predominant source of androgen excess in PCOS, the time course of androgen suppression has not previously been examined. In most studies, T was not measured until 1 month after commencing therapy, by which time, mean levels had fallen by 48–77% (25, 26, 27, 28). In the present study, T levels decreased by 20% as early as 4 h after administration of the GnRH antagonist, falling into the normal range at 8 h. In the PCOS patients treated with the GnRH antagonist for 7 days, no further inhibition of T was observed. The degree of T suppression was similar in normal and PCOS subjects. The failure of T levels, in either normal or PCOS women, to suppress to the range of oophorectomized women observed with prolonged GnRH agonist therapy (25, 26, 27, 28) may reflect the shorter duration of antagonist administration. Alternatively, it may be that the Nal-Glu GnRH antagonist used in this study is a less potent suppressor of ovarian synthesis/secretion than its agonist counterparts.
Unlike T, the DHEAS response differed between normal and PCOS women, with mean levels decreasing by 30 and 20% at 12 and 20 h, respectively, post antagonist in the controls, whereas they were unchanged in the PCOS subjects. Given that the antagonist was always given at 1200 h, it is possible that the differential DHEAS response may reflect the presence of a circadian rhythm in DHEAS secretion in the normal women that is lost in PCOS.
From this study, we conclude that the overall quantity of GnRH secreted is similar in PCOS and normal women. Thus, it seems that it is the frequency of GnRH stimulation of the pituitary that underlies the increased LH/FSH ratio in this disorder. In addition, these data imply that it is enhanced pituitary sensitivity that is responsible for the elevated LH pulse amplitude in PCOS. The fall in T levels, to the normal range, with GnRH receptor blockade confirms the LH dependence of androgen secretion in PCOS.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received December 31, 1997.
Revised March 4, 1998.
Accepted March 23, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-subunit from the gonadotrope
by gonadotropin-releasing hormone (GnRH): evidence from the use of two
GnRH antagonists. J Clin Endocrinol Metab. 70:328–335.[Abstract]
-carboxymethylmercaptotestosterone-bovine serum albumin conjugate:
measurement of testosterone in male plasma without chromatography. Steroids. 28:101–109.[CrossRef][Medline]
This article has been cited by other articles:
 |
|
 |
|
  S. S. Srouji, Y. L. Pagan, F. D'Amato, A. Dabela, Y. Jimenez, J. G. Supko, and J. E. Hall Pharmacokinetic Factors Contribute to the Inverse Relationship between Luteinizing Hormone and Body Mass Index in Polycystic Ovarian Syndrome J. Clin. Endocrinol. Metab., April 1, 2007; 92(4): 1347 - 1352. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S.K. Blank, C.R. McCartney, and J.C. Marshall The origins and sequelae of abnormal neuroendocrine function in polycystic ovary syndrome Hum. Reprod. Update, July 1, 2006; 12(4): 351 - 361. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. L. Pagan, S. S. Srouji, Y. Jimenez, A. Emerson, S. Gill, and J. E. Hall Inverse Relationship between Luteinizing Hormone and Body Mass Index in Polycystic Ovarian Syndrome: Investigation of Hypothalamic and Pituitary Contributions J. Clin. Endocrinol. Metab., April 1, 2006; 91(4): 1309 - 1316. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Gill, J. L. Sharpless, K. Rado, and J. E. Hall Evidence That GnRH Decreases with Gonadal Steroid Feedback but Increases with Age in Postmenopausal Women J. Clin. Endocrinol. Metab., May 1, 2002; 87(5): 2290 - 2296. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. R. Palmert, C. M. Gordon, A. I. Kartashov, R. S. Legro, S. J. Emans, and A. Dunaif Screening for Abnormal Glucose Tolerance in Adolescents with Polycystic Ovary Syndrome J. Clin. Endocrinol. Metab., March 1, 2002; 87(3): 1017 - 1023. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Gill, A. E. Taylor, K. A. Martin, C. K. Welt, J. M. Adams, and J. E. Hall Specific Factors Predict the Response to Pulsatile Gonadotropin-Releasing Hormone Therapy in Polycystic Ovarian Syndrome J. Clin. Endocrinol. Metab., June 1, 2001; 86(6): 2428 - 2436. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. D. Veldhuis, S. M. Pincus, M. C. Garcia-Rudaz, M. G. Ropelato, M. E. Escobar, and M. Barontini Disruption of the Joint Synchrony of Luteinizing Hormone, Testosterone, and Androstenedione Secretion in Adolescents with Polycystic Ovarian Syndrome J. Clin. Endocrinol. Metab., January 1, 2001; 86(1): 72 - 79. [Abstract] [Full Text]
|
|
|
|
|
|
F. J. Hayes, S. B. Seminara, S. DeCruz, P. A. Boepple, and W. F. Crowley Jr. Aromatase Inhibition in the Human Male Reveals a Hypothalamic Site of Estrogen Feedback J. Clin. Endocrinol. Metab., September 1, 2000; 85(9): 3027 - 3035. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
).
The degree of gonadotropin inhibition was similar in both groups at all
doses. A dose-dependent effect was apparent for LH but not FSH.
