17α-Hydroxyprogesterone Responses to Leuprolide and Serum Androgens in Obese Women with and without Polycystic Ovary Syndrome after Dietary Weight Loss 1

Abstract

Insulin resistance and increased ovarian cytochrome P450c17 activity (i.e. increased 17-hydroxylase and, to a lesser extent, increased 17,20-lyase) are both features of the polycystic ovary syndrome (PCOS). Evidence suggests that hyperinsulinemia may stimulate ovarian P450c17 activity in obese women with PCOS. We hypothesized that weight loss would decrease serum insulin and P450c17 activity in PCOS. Therefore, we measured serum ste-roid concentrations and 17-hydroxyprogesterone responses to leu-prolide administration and performed oral glucose tolerance tests before and after 8 weeks of a hypocaloric diet in 12 obese women with PCOS (PCOS group) and 11 obese women with normal menses (con-trol group). Serum insulin decreased in both groups. In the PCOS group, basal serum 17-hydroxyprogesterone decreased from 4.2 0.6 to 3.0 0.5 nmol/L (P 0.05), and leuprolide-stimulated peak serum 17-hydroxyprogesterone decreased from 14.9 2.6 to 8.9 0.8 nmol/L (P 0.025). Serum testosterone decreased from 2.47 0.52 to 1.56 0.33 nmol/L (P 0.05), and free testosterone decreased from 9.03 1.39 to 5.95 0.50 pmol/L (P 0.02). None of these values changed in the control group. Serum sex hormone-binding globulin increased by 4.5-and 3-fold in the PCOS (P 0.003) and control (P 0.007) groups, respectively. We conclude that dietary weight loss decreases ovarian P450c17 activity and reduces serum free testosterone concentrations in obese women with PCOS, but not in obese ovulatory women. The changes in women with PCOS may be related to a reduction in serum insulin. (J Clin Endocrinol Metab 82: 556-560, 1997) T HE POLYCYSTIC ovary syndrome (PCOS) affects approximately 6% of women of reproductive age and is characterized by anovulation and hyperandrogenism (1). A frequent feature of PCOS is insulin resistance and a compensatory hyperinsulinemia, and obese and nonobese women with PCOS are more insulin resistant and hyperin-sulinemic than age-and weight-matched normal women (2-13). Hyperinsulinemia plays a pathogenic role in producing the hyperandrogenism of PCOS (14, 15) by both increasing ovarian androgen production (16-18) and decreasing serum sex hormone-binding globulin (SHBG) (18-21). Hence, serum free testosterone declines in women with PCOS when circulating insulin is reduced by either pharmacological intervention (17, 18, 22) or diet (23, 24). Hyperinsulinemic insulin resistance is a salient feature of adolescent girls with hyperandrogenism (13), supporting the idea that hyperin-sulinemia plays an early and central role in the pathogenesis of PCOS. 17-Hydroxylase and 17,20-lyase are both functions of a single protein, cytochrome P450c17. Many women with PCOS also have increased ovarian 17-hydroxylase activity and, to a lesser extent, increased 17,20-lyase activity (25, 26). A hallmark of increased ovarian P450c17 activity is an exaggerated serum 17-hydroxyprogesterone response to stimulation by GnRH agonists (25-29). Hyperinsulinemia may stimulate ovarian cytochrome P450c17 activity in PCOS. We demonstrated that administration of the insulin-sensitizing agent metformin to obese women with PCOS lowers serum insulin, decreases ovarian P450c17 activity toward normal, decreases serum free tes-tosterone, and increases serum SHBG (18). If hyperinsulinemia stimulates ovarian P450c17 activity in obese women with PCOS, then nonpharmacological methods for lowering serum insulin, such as weight loss, should also be associated with an improvement in ovarian P450c17 activity. Obesity is an insulin-resistant state (30), and weight loss in obese individuals is accompanied by improved insulin sensitivity and a reduction in serum insulin (31-33). To test this possibility, we assessed ovarian P450c17 activity by a leuprolide stimulation test in obese women with and without PCOS before and after dietary weight loss. Subjects and Methods Subjects Twelve women with PCOS (PCOS group) and 11 women with normal menses (control group) were studied. All women were obese [body mass index (BMI), 27.5 kg/m 2 ] and 18-35 yr old. None had taken any medications for 2 months or more before the study, and none had diabetes mellitus. PCOS was defined by oligomenorrhea (6 menstrual periods in the last year) and hyperandrogenemia (elevated serum free testosterone).