Girls with high levels of the male hormone testosterone often develop polycystic ovary syndrome (PCOS) as adults. Women with PCOS often have irregular menstrual periods, excess facial and body hair, and weight gain. Women with PCOS also have difficulty becoming pregnant. Some, girls with high levels of male hormone will develop normal hormone levels as they grow up. Most girls continue to have high levels of male hormone as adults. In addition, girls with elevated levels of male hormones often have lower fertility rates in adulthood. In this study the investigators will aim to discover the effect of 7 days of estrogen and progesterone on GnRH pulses in girls and women with the goal of understanding how and why some girls and women have higher levels of male hormone and the causes of PCOS. If investigators understand the causes of these disorders, they may be able to better treat them and perhaps even learn how to prevent the development of PCOS.

Adolescent hyperandrogenemia (excess androgen production) occurring before or during early puberty appears to be a precursor to adult polycystic ovary syndrome (PCOS). PCOS affects about 6% of women of childbearing age in the United States. Those who suffer from this disorder often experience irregular menstrual periods, excess facial and body hair, and weight gain. PCOS is also a leading cause of infertility. Women with PCOS often report irregular menstrual cycles as adolescents. A study of adolescents with menstrual irregularities showed that some subjects normalize endocrine function as they mature, while a majority maintained hyperandrogenism in conjunction with high levels of luteinizing hormone (LH) and polycystic ovaries. In addition, girls with high levels of serum androgens often have lower fertility rates in adulthood.

We propose that adult PCOS, and perhaps adolescent hyperandrogenemia, are due in part to dysregulation of pituitary and ovarian hormones. Synthesis and secretion of LH and follicle stimulating hormone (FSH) are primarily regulated by gonadotropin releasing hormone (GnRH). Both LH and FSH are secreted by the same gonadotrope cell, and the frequency of stimulation of this cell by GnRH in part determines which hormone is released. In primates, rapid GnRH frequencies (approx. 1 pulse/ hour) favor LH secretion whereas slower GnRH stimuli (1 pulse/ 3 hours or less) favor FSH release. In normal women, the cyclical rise and fall in hormone levels control follicular maturation and ovulation. Early studies showed an initial predominance of FSH in the follicular phase, with a subsequent rise in estradiol (E2). In the late follicular phase, LH increases as a consequence of increased GnRH secretion. Following ovulation, rising levels of E2 and P then reduce GnRH pulse frequency, allowing a rise in FSH for the next cycle of follicular maturation.

One feature of adult PCOS is increased mean serum levels of LH and increased LH pulse frequency, presumably due to increased stimulation of the pituitary by excess hypothalamic secretion of GnRH. Since women with PCOS maintain high levels of LH and low levels of FSH, follicle maturation and ovulation do not occur normally. Girls with hyperandrogenemia in adolescence also have an increased frequency of LH pulses when compared to age matched controls.

If hyperandrogenemic adolescents could be treated effectively before or during pubertal maturation, development of clinical PCOS as an adult could potentially be avoided. One proposed cause of both hyperandrogenemia and PCOS is a defect in GnRH pulse modulation, which normally happens as puberty progresses. GnRH is secreted by a part of the brain called the hypothalamus. In normal pubertal maturation the increase in GnRH pulse secretion during sleep stimulates LH and ovarian E2 and P secretion. Feedback of these hormones reduces GnRH pulses during daytime hours, initiating cycles of ovarian-hypothalamic feedback regulation which mature into the patterns seen in normal ovulatory cycles. Recent studies have shown that E2 and P can slow LH pulses in adult women with PCOS, but higher concentrations of P are needed to inhibit LH pulse frequency. If hypothalamic (GnRH pulse generator) sensitivity to inhibition by P is reduced during pubertal maturation, the low levels of P present during the initial development of ovarian cyclicity may not be adequate to suppress GnRH/LH pulse secretion. This could lead to LH excess and relative FSH deficiency. Administering oral doses of P in early adolescence may compensate and restore normal ovarian-hypothalamic feedback. In turn, increasing amounts of naturally secreted P in subsequent cycles could eventually normalize the system.

The long term goal of this line of investigation is to determine if E2 and P treatment of adolescents with hyperandrogenemia can slow the GnRH pulse generator to promote FSH production and the advent of normal menstrual cycles. As an initial step we propose to determine if the GnRH pulse generator is relatively insensitive to E2 and P inhibition in hyperandrogenemic adolescent girls.

Inclusion Criteria:

• Girls ages 8 to 18
• Hyperandrogenemic (testosterone level > 0.4 ng/mL and/or hirsutism)
• Normal screening labs (with exception of the expected hormonal abnormalities inherent in hyperandrogenemia)

Exclusion Criteria:

• Abnormal screening labs (with exception of the expected hormonal abnormalities inherent in hyperandrogenemia)
• Congenital adrenal hyperplasia.
• Hemoglobin <12 mg/dL or hematocrit < 36% (Subjects will be offered the opportunity to take iron supplementation for 60 days if their hematocrit is slightly low (33-36%) (suggestive of iron deficiency anemia) and will then return for retesting of their hemoglobin/hematocrit.)
• Weight < 31 kg
• History of peanut allergy, deep venous thrombosis, breast cancer, endometrial cancer, or cervical cancer
• On hormonal medications (including oral contraceptive pills) or on medications known to affect the reproductive axis within 3 months of the study
• Pregnant or breastfeeding
• Participation in a research study within the past 30 days that involved taking a study drug.
• Participation in a research study that involved taking up to or greater than 473 ml's of blood within the past 60 days.
• Cigarette smoking
• History of surgery that required bedrest within the past 30 days
• Family history of hypercoagulability or unexplained thromboembolic disease (not in setting of bedrest, surgery, or malignancy)