| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Division of Endocrinology (R.P., L.P., A.G.), Department of Internal Medicine, Centre for Applied Biomedical Research, Sant’Orsola-Malpighi Hospital, University Alma Mater Studiorum of Bologna, and Day-Surgery Center (P.P., G.E.C.), Gynepro Medical, 40138 Bologna, Italy
Address all correspondence and requests for reprints to: Renato Pasquali, M.D., Division of Endocrinology, Department of Internal Medicine, Sant’Orsola-Malpighi Hospital, University Alma Mater Studiorum, Via Massarenti 9, 40138 Bologna, Italy. E-mail: renato.pasquali{at}unibo.it.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Objective: Our objectives were to quantify the prevalence of PCOS with FOH and to evaluate whether the presence of FOH may distinguish different clinical and biochemical phenotypes.
Design, Setting, and Participants: We conducted an observational study at an academic hospital that included 148 PCOS women and 22 healthy age-matched normal-weight control women.
Main Outcome Measures: A hormone profile was taken at baseline and in response to 1–24ACTH and to a GnRH agonist, buserelin, administered during dexamethasone suppression.
Results: Based on the data obtained in the control subjects, the PCOS patients were divided into two groups, one with a normal (NR-PCOS, n = 78) and one with a high 17-hydroxyprogesterone response (HR-PCOS, n = 70) to buserelin. The two groups of PCOS subjects had similar anthropometric parameters and clinical signs of hyperandrogenism. Age and body weight at menarche were significantly lower and higher, respectively, in the HR-PCOS group than the NR-PCOS group. Moreover, the HR-PCOS group had higher basal testosterone (P < 0.001), free androgen index (P < 0.01), 17-hydroxyprogesterone (P < 0.05), estrogens (P < 0.05), area under the curve for insulin (insulinAUC) (P < 0.05), and C-peptideAUC (P < 0.01) and lower insulin sensitivity (as composite insulin sensitivity index) (P < 0.05) than the NR-PCOS group. The response of 17-hydroxyprogesterone to 1–24ACTH (as percent variation) was lower in the HR-PCOS group with respect to the NR-PCOS group (P < 0.05), whereas the response of cortisol, androstenedione, and dehydroepiandrosterone was similar. Finally, the HR-PCOS group had lower percent suppression of androstenedione (P < 0.001) and 17-hydoxyprogesterone (P < 0.05) to dexamethasone. In a multiple regression model applied in all PCOS women, insulinAUC but not androgens or markers of insulin resistance predicted the 17-hydroxyprogesterone response to buserelin to a highly significant extent (t = 3.269; P < 0.01).
Conclusions: This study indicates that the paradigm that FOH is a specific feature of the PCOS status can no longer be sustained. We have shown that women with an exaggerated 17-hydroxyprogesterone response to a GnRH agonist, buserelin, are characterized by more severe hyperandrogenemia, glucose-stimulated ß-cell insulin secretion, and worse insulin resistance than those without evidence of FOH. Our data may be consistent with the hypothesis that excess insulin may represent a candidate factor responsible for FOH in these women, through the overactivation of the cytochrome P450 17
-hydroxylase/17,20-lyase (CYP17) enzyme pathway.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
4-pathways, in the ovaries (1, 2, 3, 4, 5). Accordingly, it has been proposed that the prominent abnormality in most women with polycystic ovary syndrome (PCOS) is represented by an excessive 17-hydroxyprogesterone response to GnRH agonists, reflecting ovarian androgen overproduction. Although this concept derived from studies performed in relatively small groups of homogeneous women with PCOS, it nonetheless leads to important modifications in the pathophysiological understanding of the disorder. Many other studies have successfully used this test to prove ovarian dysfunction in PCOS women, and the original concept has been further substantiated by the finding that treatment with lifestyle intervention (6) and insulin sensitizers, such as metformin (7, 8), significantly reduced or normalized the 17-hydroxyprogesterone response to GnRH agonists in different groups of affected women. As a consequence, in the last decade, we routinely applied this test not only to perform the diagnosis of hyperandrogenic disorders but also to increase our pathophysiological knowledge, which is potentially relevant for therapeutic strategies. However, in our extensive clinical experience, we found that not all patients with well-defined PCOS presented with an exaggerated 17-hydroxyprogesterone response to GnRH agonists, being conversely characterized by a normal response. This therefore makes it difficult to classify these patients according to a common pathophysiological dysregulation, as previously suggested.
With this background, we therefore carried out this study in a large group of women with PCOS to evaluate 1) the prevalence of subjects with an exaggerated or normal 17-hydroxyprogesterone response to a GnRH agonist, buserelin, and 2) whether the presence of FOH may distinguish different clinical and biochemical phenotypes, suggesting specific pathophysiological mechanisms.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
We prospectively analyzed 148 consecutive PCOS women in reproductive age (mean age, 24.9 ± 6.0 yr; range, 15–49 yr) attending the Division of Endocrinology of the Sant’Orsola-Malpighi Hospital of Bologna, Italy, as outpatients from 2000–2005. The diagnosis of PCOS was performed using the presence of chronic anovulation (progesterone levels in the luteal phase of the last cycle before recruitment not exceeding 2 ng/ml), mild to moderate (women with at least one cycle every 45–60 d in the last 7 months) or severe oligomenorrhea (three or fewer cycles in the last 7 months) or amenorrhea (no cycles in the last 7 months), and clinical (hirsutism, defined by the Ferriman-Gallwey score 
8) (9) and/or biochemical signs of hyperandrogenism (elevation of both total and free testosterone levels), after exclusion of other causes of hyperandrogenism such as Cushing’s syndrome, nonclassical adrenal hyperplasia, androgen-secreting tumors, drug-induced androgen excess, and hyperprolactinemia, according to the 1990 National Institutes of Health proposed criteria (10). Only two women were aged over 40 (43 and 49 yr, respectively). Their clinical data and hormone levels (based on repeated blood samples taken within 6 months) did not support a menopausal status. One of them had severe oligomenorrhea and the second one mild to moderate oligomenorrhea (all had hirsutism, hyperandrogenemia, and polycystic ovaries). None of the included subjects had thyroid dysfunction, abnormal prolactin levels, type 2 diabetes, or cardiovascular, renal, or liver dysfunction based on clinical examination and routine laboratory findings or had taken any medication in the 3 months preceding clinical and laboratory evaluations. Twenty-two healthy age-matched (age, 26.5 ± 3.6 yr; range, 22–40 yr) normal-weight women were also included in the study, as a control reference group for basal hormone and metabolic parameters and functional test data. All women in the control group had normal menses and were ovulatory, and none had hirsutism, acne, alopecia, or acanthosis nigricans. Informed consent was obtained from all PCOS and control subjects before participating in the study, which was approved by the local ethics committee. All women underwent anthropometric, clinical, and laboratory evaluations, as described below.
Anthropometry and clinical evaluations
Anthropometry included the measurement of height, weight, and waist and hip circumference according to standardized procedures, as previously described (11). Overweight or obesity states were defined as a body mass index (BMI) ranging from 25.1–30 kg/m2 or greater than 30 kg/m2, respectively, and normal weight as a BMI ranging from 18–25 kg/m2 (12). Other than for hirsutism, evaluated by the Ferriman-Gallwey score, and used for the diagnosis of PCOS, subjects were also evaluated for androgenic alopecia and acne, for other clinical signs of hyperandrogenism, and for acanthosis nigricans, as previously reported (11).
A questionnaire regarding personal and family history was given to each subject enrolled in the study and was filled in at home with the help of the subject’s parents, if possible, and, finally, of one of the investigators (A.G.). The questionnaire addressed body weight at birth, age of menarche, body weight at menarche and familiality for diabetes and obesity, which was considered positive only if at least one first-degree relative was or had been affected. The reliability of the data relating to birth weight, obtained in 120 women with PCOS and in 20 of the control group, was assessed by the personal pediatric record, if available, as previously described (11). Dietary intake was evaluated by the dietician, and physical activity was investigated as previously described (11, 13).
Functional tests and laboratory evaluations
In the PCOS women with mild to moderate oligomenorrhea and in the control group, the protocol was applied starting from d 2 or 3 of the menstrual cycle to complete it on d 8 or 9. On the contrary, the test sequence was performed randomly in PCOS women presenting with severe oligomenorrhea or amenorrhea. Human chorionic gonadotropin was also measured in each woman at baseline to exclude pregnancy. Blood samples were drawn and all tests were performed in all women from 0800–0900 h after a 12-h overnight fast, provided they had followed a 3-d diet containing at least 250–300 g carbohydrates per day.
On the first day, after baseline blood samples for hormonal and metabolic parameters had been obtained, an oral glucose tolerance test (OGTT) (75 g Curvosio; Sclavo, Cinisello Balsamo, Italy) was performed, taking blood samples after 30, 60, 90, 120, and 180 min for glucose determination and after 60, 120, and 180 min for insulin and C-peptide determinations. Samples for hormone assay were immediately chilled on ice and centrifuged, and serum aliquots were collected and frozen at –20 C until assayed.
On the second day, a 1–24ACTH stimulation test (Synacthen, 250 µg iv) was performed, and samples for cortisol, dehydroepiandrosterone (DHEA), androstenedione, and 17-hydroxyprogesterone determinations were obtained at baseline and 30 and 60 min after stimulation. These samples were immediately chilled on ice and centrifuged, and plasma aliquots were collected and frozen at –20 C until assayed. The same day, starting from 1200 h, 0.5-mg dexamethasone tablets were administered every 6 h for 5 consecutive days to suppress adrenal steroidogenesis. On d 4 of the dexamethasone suppression test, blood samples were taken again in the morning (from 0800–0830 h) to measure the same hormone concentrations. One additional blood sample was obtained for androgen and estrogen measurements and two additional blood samples, with a 15-min interval, to measure gonadotropins (LH and FSH). Buserelin, a GnRH agonist (Superfact, 1 mg sc) was then administered, taking blood samples for gonadotropins after 30 and 60 min thereafter. Twenty-four hours later (from 0800–0830 h), with women still on dexamethasone suppression (the last pill was taken at 0600 h), two additional samples were obtained with a 15-min interval to measure gonadotropins, androgens, and estrogens.
Assays
Plasma glucose levels were determined by the glucose oxidase technique immediately after the blood drawing. Hormones and lipids, including total cholesterol, high-density lipoprotein (HDL) cholesterol, and triglycerides, were measured as previously described (14). The free androgen index (FAI) was calculated as the ratio between total testosterone and SHBG, according to Vermeulen et al. (15). To investigate insulin sensitivity in basal condition, the homeostasis model assessment for insulin resistance (HOMA-IR) was calculated (16); in addition, the composite insulin sensitivity index (ISIcomposite) was determined from the results of the OGTT (17). The intraassay coefficients of variation in our laboratory were 3.0% for insulin, 10% for C-peptide, 4.8% for LH, 1.9% for FSH, 7.0% for testosterone, 6.0% for androstenedione, 6.0% for DHEA, 5.9% for DHEA sulfate (DHEAS), 13% for 17-hydroxyprogesterone, 11% for progesterone, 5.6% for estradiol, 9% for estrone, 10% for cortisol, and 6.5% for SHBG.
Statistics and data analysis
Glucose, insulin, and C-peptide responses to the OGTT were expressed as area under the curve (AUC), which was calculated by the trapezoidal method. Hormone responses to the dexamethasone suppression test and to the buserelin stimulation test were calculated as percent changes vs. baseline values. For the 1–24ACTH stimulation test, hormone response was calculated by percent variation of peak values with respect to basal concentrations. For the buserelin test, the arithmetic mean of the two basal gonadotropin values and those of hormones measured 24 h later were used to calculate percent changes. The response of LH and FSH to buserelin was also evaluated by the AUC, which was calculated by the trapezoidal method. The cutoff value used to define the response of 17-hydroxyprogesterone to buserelin in PCOS women was calculated from the antilogarithm of mean + 1.96 SD of 17-hydroxyprogesterone values enriched 24 h after buserelin administration in the control group. Normal distribution and homoscedasticity of continuous variables were tested by means of the Kolmogorov-Smirnov (18) and the Levene tests (19). Variables that did not fulfill these tests were log-transformed before analysis. The data were compared between the groups by ANOVA. Because of a significantly different mean age, all comparisons between PCOS groups and normal-weight controls were performed by ANOVA, after adjusting for BMI values. Simple correlation analyses were used to relate the response of 17-hydroxyprogesterone to buserelin (expressed as percent variation) with androgen and metabolic parameters in all PCOS women. A multiple regression model was also used to evaluate interactions between all independent variables with 17-hydroxyprogesterone response to buserelin.
All statistical analyses were performed using StatView software (Acabus Concepts Inc., Berkeley, CA). All data are expressed as mean ± SD or SEM, when indicated. P values < 0.05 were regarded as statistically significant.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
According to the cutoff value of 17-hydroxyprogesterone response to buserelin observed in our control subjects, PCOS patients were divided into two groups, one presenting with a normal response (NR-PCOS, n = 78) and the other with a high response (HR-PCOS, n = 70) (Fig. 1
).
|
The general characteristics of the HR-PCOS and NR-PCOS subjects are reported in Table 1. There were no significant differences in any anthropometric parameters (with the exception of higher values of hip circumference measurement in the NR-PCOS group) or in the prevalence of normal-weight or overweight-obese individuals between the two groups. As expected, all anthropometric parameters (except height) were significantly higher in both PCOS groups with respect to the control group.
|
). Dietary intake was not significantly different between the two PCOS groups, although it was significantly higher than that of the control group (P < 0.01) (data not shown). Physical activity scores were similar among all the groups (data not shown). Basal sex hormones and SHBG
Compared with the control subjects, both PCOS groups had significantly higher basal testosterone, FAI, androstenedione, and LH and lower SHBG and cortisol levels, whereas DHEAS and progesterone did not differ. In the HR-PCOS women, but not in the NR-PCOS group, FSH levels were lower than in the control group. Significantly higher basal testosterone, FAI, and 17-hydroxyprogesterone concentrations were found in the HR-PCOS subjects compared with the NR-PCOS women (Table 2). In addition, HR-PCOS women had higher estrogen (estradiol and estrone) concentrations than NR-PCOS women and the control subjects, without any significant difference between the latter two groups.
|
Both PCOS groups had higher fasting and stimulated insulin and C-peptide concentrations compared with the control group but similar fasting glucose and glucoseAUC values (Table 3 and Fig. 2). Moreover, they were characterized by higher HOMA-IR and lower ISIcomposite values and had lower HDL-cholesterol and higher triglyceride levels (Table 3). However, with respect to NR-PCOS women, the HR-PCOS group had significantly higher insulinAUC and C-peptideAUC and lower ISIcomposite, whereas fasting glucose, insulin, and C-peptide, HOMA-IR, and lipids were similar (Table 3 and Fig. 2).
|
|
Basal and stimulated levels of each hormone are reported in Fig. 3. Cortisol response (as percent variation of peak) had a nonsignificant tendency to be higher in the HR-PCOS and NR-PCOS groups with respect to the control group, whereas the response of 17-hydroxyprogesterone was significantly lower in the HR-PCOS group than in the NR-PCOS and the control groups. The 17-hydroxyprogesterone response was below the range for late-onset congenital adrenal hyperplasia in all PCOS women, thus confirming appropriate patient selection. The responses (as percent variation of peak values) of androstenedione and DHEA were similar between the groups, although their stimulated concentrations were still higher than in the controls.
|
Cortisol suppression was similar in all groups (>85%), and no woman had suppression lower than 80% (Table 4). No significant difference was also found in DHEA and DHEAS percent suppression between HR-PCOS and NR-PCOS, although in both groups, percent suppression of DHEAS tended to be significantly lower than in the controls. Conversely, androstenedione percent suppression was significantly lower in the HR-PCOS women with respect to the NR-PCOS group, without any significant difference between the PCOS groups and controls. Finally, the HR-PCOS group was characterized by significantly lower 17-hydroxyprogesterone suppression in comparison with both NR-PCOS women (P < 0.05) and controls (P < 0.01) (Table 4).
|
All groups were characterized by a rapid increase in LH levels, with peak values after 30 min in the HR-PCOS group and after 60 min in the other groups, whereas peak levels for FSH, which conversely occurred after 24 h, were similar in all groups (Fig. 4). LHAUC was similar in both HR-PCOS and NR-PCOS women but higher in both groups with respect to controls. FSHAUC values were similar in all groups.
|
Twenty-four hours after buserelin stimulation, androstenedione, progesterone, and testosterone were significantly higher in HR-PCOS subjects than in NR-PCOS and control subjects (Fig. 4). For all hormones, the percent variation was, however, similar, with a trend toward a higher progesterone response in the HR-PCOS women with respect to the NR-PCOS group. Estradiol levels rose significantly more in the HR-PCOS women than the NR-PCOS and control groups, but their percent variation was not significantly different.
Relationship between 17-hydroxyprogesterone response to buserelin, clinical, hormonal, and metabolic parameters
The response (as percent variation) of 17-hydroxyprogesterone to buserelin was significantly correlated with both pre-buserelin test hormone levels (r = 0.441; P < 0.001) and pre-1–24ACTH test (r = 0.35; P < 0.01). Moreover, the percent variation of 17-hydroxyprogesterone was correlated with LHAUC (r = 0.303; P < 0.001). Significant correlations were also found with testosterone (r = 0.382; P < 0.001) and FAI (r = 0.372; P < 0.01) and with insulinAUC (r = 0.482; P < 0.01), C-peptideAUC (r = 0.473; P < 0.01), and ISIcomposite (r = –0.487; P < 0.01) values. By contrast, no correlations were found with BMI, waist circumference, or waist-to-hip ratio values. Finally, no correlation was found between the androgen (17-hydroxyprogesterone, testosterone, and androstenedione) response to buserelin and the corresponding androgen response to 1–24ACTH test or with suppression of each androgen after dexamethasone administration. In a multiple regression model with the percent variation of 17-hydroxyprogesterone as a dependent variable and testosterone, FAI, LHAUC, insulinAUC, C-peptideAUC, and ISIcomposite as independent variables, the correlation was still significant for insulinAUC, regardless of whether androgens (testosterone and FAI) were included (t = 3.269; P < 0.01) or excluded (t = 3.646; P < 0.001) from the analysis.
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Our study was performed in a large cohort of well-defined PCOS women, based on original National Institutes of Health criteria (10) to exclude potentially heterogeneous phenotypes, covering a wide range of BMI values. In addition, our protocol was very similar to that used in original studies (1, 2), the only difference being the use of buserelin instead of nafarelin. Previous studies have, however, shown that the specific potency of each available GnRH agonist in the acute stimulation of the pituitary-gonadal axis is similar (20), as is their inhibiting capacity of gonadotropin and sex steroids after 3 months of treatment (21). In addition, previous studies performed in women with PCOS have found a higher than normal 17-hydroxyprogesterone response to GnRH agonist testing using nafarelin (1), buserelin (22, 23), or leuprorelin acetate (24). In one study using the same doses of buserelin that we did (1 mg, sc), the 17-hydroxyprogesterone peak values reported were consistent with those we observed in the present study (23). It is therefore unlikely that the use of buserelin instead of nafarelin (or leuprolide acetate) may be responsible for differences in the testing evaluation of FOH in our PCOS women.
According to previous reports (1, 2, 22, 23, 24), the LH response to buserelin in all PCOS women was significantly higher than that observed in the normal-weight control group, without any difference between the HR-PCOS and the NR-PCOS groups. This suggests that different 17-hydroxyprogesterone responses in the two PCOS groups are partly independent of the endogenous gonadotropin concentrations recorded during the test. Moreover, the exaggerated 17-hydroxyprogesterone response in the HR-PCOS women was approximately 3-fold higher than that in the NR-PCOS group. This makes it unlikely that this response may depend on higher basal values of 17-hydroxyprogesterone, as confirmed by the values of the variance (approximately 10%) explained by the correlation coefficient we found between basal and buserelin-stimulated hormone values. Accordingly, the marked increase of other androgens, such as androstenedione and testosterone, appears to be acceptably dependent, at least in part, on the higher baseline values, although a specific overactivation of the 4-pathway beyond progesterone, which appears to be relatively independent of an increased LH drive, is likely to occur in HR-PCOS subjects. This is emphasized by the higher increase in estradiol response in the same groups, occurring regardless of the fact that pretest levels were similar to those observed in both NR-PCOS and control subjects, which suggests an increased peripheral conversion from the larger amount of testosterone elicited by buserelin stimulation in the HR-PCOS women. A direct effect of FSH on estrogen production can probably be excluded, because all groups had similar FSH responses to buserelin stimulation. Finally, it is unlikely that the abnormal 17-hydroxyprogesterone response to buserelin we found in the HR-PCOS women may depend on supraphysiological stimulation. In fact, McCartney et al. (25) found a higher 17-hydroxyprogesterone response in a group of PCOS women with respect to controls, even after near-physiological LH infusion, which reinforces the concept that hyperresponsiveness of 17-hydroxyprogesterone to buserelin may represent a genuine alteration in the HR-PCOS women with specific pathophysiological relevance.
In contrast to previous studies (1, 2), we found that, unlike other androgens, the suppression of androstenedione and 17-hydroxyprogesterone by dexamethasone was significantly lower in the HR-PCOS group compared with their NR-PCOS counterpart. By contrast, the 17-hydroxyprogesterone response to 1–24ACTH was significantly higher in the former than in the latter, which may be consistent with some dysregulation of the cytochrome P450 17-hydroxylase/17,20-lyase (CYP17) adrenal tissue, as previously suggested (3, 5). Nonetheless, we could not find any significant correlation between 17-hydroxyprogesterone, androstenedione, and testosterone responses to buserelin and that of each hormone after the standard 1–24ACTH test or their suppression after dexamethasone suppression, which partially contrasts with this hypothesis, unless the presence of different pathophysiological mechanisms responsible for tissue-specific dysregulation of the ovarian and adrenal CYP17 enzyme activities are suggested. Rather, it is likely that a specific intrinsic defect within the thecal-interstitial androgen-secreting cells is present in PCOS women overresponding in 17-hydroxyprogesterone production (1). This is supported by other studies that failed to find any difference in the adrenal CYP17 enzyme activity between hyperandrogenized PCOS women and healthy controls (26). However, some studies have suggested that excess androgen production after standard (26, 27) or low-dose (28) 1–24ACTH test stimulation may represent a generalized alteration of synthesis by adrenocortical control, being in any case present also in women with idiopathic hirsutism (27).
As previously suggested (5), the findings reported here are consistent with the hypothesis that insulin may represent the candidate factor amplifying LH effectiveness on 17-hydroxyprogesterone overstimulation in the HR-PCOS group. In fact, this subgroup of women with PCOS was characterized by higher glucose-stimulated ß-cell function and worse insulin-resistant state, which were significantly correlated with the 17-hydroxyprogesterone response to buserelin stimulation. We hypothesize that insulin-mediated overactivation of the CYP17 enzyme system may be responsible for both basal and GnRH-stimulated hyperandrogenism in the HR-PCOS women, in comparison with the NR-PCOS group. This implies that mechanisms for androgen excess and insulin action on ovarian steroidogenesis are not uniform in PCOS. The CYP17 system is encoded by a single gene on chromosome 10 and is expressed in ovarian thecal cells, which are the main although not exclusive target for insulin action among ovarian components (29). The stimulatory capacity of insulin on ovarian steroidogenesis was recognized many years ago by both in vitro and in vivo studies (30). Insulin does in fact stimulate LH-induced theca cell steroidogenesis (31, 32), specifically acting on the steroidogenic acute regulatory protein and the CYP17 enzyme system, including both 17-hydroxylase and 17,20-lyase activities (33). Moreover, insulin synergizes with both IGF-I (34, 35) and estradiol (36) to increase androstenedione production. According to our findings, there are clinical studies that have shown a significant relationship between 17-hydroxyprogesterone response to GnRH analogs and indices of insulin resistance (37), which may suggest a greater androgenic response in the presence of more severe insulin resistance. In addition, studies performed to investigate the effects of improved insulin sensitivity and circulating blood insulin concentrations by lifestyle intervention (6) or insulin sensitizers (7, 8) have uniformly demonstrated a significant improvement of the response of 17-hydroxyprogestrone to GnRH analogs, reaching the normal-range level in many cases, consistent with a reduction of the CYP17 activity, although a wide range of responses were observed in these studies. Interestingly, in vitro experiments have also shown that the insulin sensitizer troglitazone dose-dependently antagonizes LH and insulin’s combined stimulation of androstenedione and testosterone production by theca cells in vitro, CYP17 gene expression, and CYP17 protein phosphorylation (38). These in vitro data were only partially confirmed by a large long-term controlled trial performed by Azziz et al. (39) who found that free testosterone decreased and SHBG increased in a dose-related fashion with troglitazone treatment administered at different doses, although neither basal gonadotropins nor the LH/FSH ratio changed with therapy. Moreover, in a subanalysis study, the same authors found that treatment with troglitazone at high doses induced a small but significant decrease in DHEAS levels. They also found a small but significant relationship between baseline DHEAS and baseline insulinAUC, although this relationship was not still significant for the same parameters measured at the end of the study. Whether changes in DHEAS may depend on decreased insulin or, alternatively, on a direct or indirect effect of troglitazone by itself on adrenal steroidogenesis is therefore still unclear.
In conclusion, this study indicates that the paradigm that the FOH is a specific feature of PCOS status can no longer be sustained. We have shown that women with an exaggerated 17-hydroxyprogesterone response to a GnRH analog, buserelin, are characterized by more severe hyperandrogenemia, glucose-stimulated ß-cell insulin secretion, and worse insulin resistance with respect to those without evidence of FOH. Our data are also consistent with the hypothesis that excess insulin may represent a candidate factor responsible for FOH in these women, through the overactivation of the CYP17 enzyme pathway.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
First Published Online September 4, 2007
Abbreviations: AUC, Area under the curve; BMI, body mass index; CYP17, cytochrome P450 17-hydroxylase/17,20-lyase; DHEA, dehydroepiandrosterone; DHEAS, DHEA sulfate; FOH, functional ovarian hyperandrogenism; HDL, high-density lipoprotein; HOMA-IR, homeostasis model assessment for insulin resistance; HR-PCOS, high-response PCOS patient; ISIcomposite, composite insulin sensitivity index; NR-PCOS, normal-response PCOS patient; OGTT, oral glucose tolerance test; PCOS, polycystic ovary syndrome.
Received April 17, 2007.
Accepted August 27, 2007.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
as the cause of polycystic ovary syndrome. Fertil Steril 53:785–791[Medline]-Hydroxyprogesterone responses to leuprolide and serum androgens in obese women with and without polycystic ovary syndrome after dietary weight loss. J Clin Endocrinol Metab 82:556–560 activity and serum free testosterone after reduction of insulin secretion in polycystic ovary syndrome. N Engl J Med 335:617–623 activity and serum androgens. J Clin Endocrinol Metab 82:4075–4079 enzyme activity in women with polycystic ovary syndrome. Eur J Endocrinol 147:473–477[Abstract]-hydroxylase/17–20 lyase (CYP17) genes in porcine theca cells. Endocrinology 141:2735–2742This article has been cited by other articles:
 |
|
 |
|
  T. Sir-Petermann, E. Codner, V. Perez, B. Echiburu, M. Maliqueo, A. Ladron de Guevara, J. Preisler, N. Crisosto, F. Sanchez, F. Cassorla, et al. Metabolic and Reproductive Features before and during Puberty in Daughters of Women with Polycystic Ovary Syndrome J. Clin. Endocrinol. Metab., June 1, 2009; 94(6): 1923 - 1930. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Hirshfeld-Cytron, R. B. Barnes, D. A. Ehrmann, A. Caruso, M. M. Mortensen, and R. L. Rosenfield Characterization of Functionally Typical and Atypical Types of Polycystic Ovary Syndrome J. Clin. Endocrinol. Metab., May 1, 2009; 94(5): 1587 - 1594. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Mortensen, D. A. Ehrmann, E. Littlejohn, and R. L. Rosenfield Asymptomatic Volunteers with a Polycystic Ovary Are a Functionally Distinct but Heterogeneous Population J. Clin. Endocrinol. Metab., May 1, 2009; 94(5): 1579 - 1586. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
, P < 0.05 for comparison between the NR-PCOS or HR-PCOS women and the normal-weight control group (performed by ANOVA, after adjusting for BMI values). C-peptideAUC was available for only 120 PCOS women (58 in the HR-PCOS group and 62 in the NR-PCOS group). To convert to SI units, see Table 2
