Polycystic ovary syndrome (PCOS) is the most common cause of anovulatory infertility. Its prevalence among women of reproductive age is estimated to be between 5-10%. PCOS is classified as a syndrome because it is a heterogeneous disorder: not all of the women with PCOS will express all of the symptoms associated with the disorder.
A diagnosis of PCOS is made if a woman has chronic problems with ovulation coupled with hyperandrogenism (excess secretion of androgens). Problems with ovulation will manifest themselves as amenorrhea (lack of menstruation) or oligomenorrhea (irregular menstruation). Hyperandrogenism may cause hirsutism, which is a masculine pattern of hair growth, and acne, because androgens have an effect on the sebaceous glands of the skin that promotes acne. Androgens may also cause hair loss on the scalp.
PCOS gets its name from the changes seen in the ovary. Polycystic ovaries are enlarged and contain multiple (greater than 10) immature follicles. The immature follicles show a relative hyperplasia of theca cells, and fewer granulosa cells. Development of these follicles is arrested well before the point of dominant follicle selection, so no LH surge can occur, and therefore, no ovulation.
There are also metabolic disturbances associated with PCOS. Frequently, women with PCOS are found to be insulin resistant. Because insulin resistance is a decreased sensitivity to insulin, this means that more insulin is necessary to achieve the same effect. For this reason, individuals who are insulin resistant have higher levels of insulin secretion or hyperinsulinemia. Because women with PCOS are insulin resistant, they are at a greater risk for developing type 2 diabetes mellitus. Many women with PCOS are also obese, which adds to their insulin resistance and risk for progression to diabetes.
Normal follicle development begins when estrogen and progesterone levels drop due to degeneration of the corpus luteum. The release from negative feedback inhibition allows a small but steady increase in FSH and LH levels that stimulates the growth phase for a group of follicles. In the first part of the follicular phase, granulosa cells respond to FSH only, while theca cells respond to LH. The hormonal interactions in the early follicular phase are shown in Figure 1.
The cause of PCOS is not at all clear, but one consistent observation is that there is an imbalance in gonadotropin production. LH secretion is elevated, while FSH secretion is the same, or even decreased. LH stimulates theca cell proliferation and secretion of androgens, but there is insufficient FSH to stimulate granulosa cells. Recall that production of estrogen by the ovary requires the activity of the enzyme aromatase that is expressed in granulosa cells. The result is high levels of androgens secreted from the ovary (hyperandrogenism), and a failure of follicle development to progress.
Figure 2 depicts how the endocrine disturbances in PCOS become part of a vicious cycle, where the abnormalities are reinforced. The androgens secreted from the ovary are converted to estrogen because certain body tissues (in particular, adipose tissue) express aromatase. This continuous level of estrogen causes abnormal feedback regulation of gonadotropin secretion, such that LH secretion continues to be high relative to FSH secretion. Hyperinsulinemia contributes to the problem because insulin stimulates ovarian androgen production.
If a woman is not seeking to get pregnant, PCOS is often treated with hormonal contraceptives. Typically, hormonal contraceptives contain a low dose of estrogen and progesterone, and are taken for 3 weeks, with one week off for a "withdrawal bleed". The estrogen and progesterone act to restore normal negative feedback regulation and lower LH secretion. This is often sufficient to reduce hyperandrogenism and its associated symptoms.
Hormonal contraceptives also provide regular menstrual cycles, and prevent excessive endometrial proliferation. In untreated PCOS, the uterus experiences high levels of estrogen, but no progesterone, because the cycle never advances to the luteal phase. Constant estrogen alone can cause excessive endometrial proliferation, and may increase a woman's risk for the development of endometrial cancer.
If the woman is seeking to get pregnant, the first line of therapy is treatment with clomiphene. Clomiphene is a selective estrogen receptor modulator (SERM), which means that it binds to the estrogen receptor, and acts as either an agonist or an antagonist depending on the tissue. In the hypothalamus and anterior pituitary, clomiphene acts as an estrogen antagonist. It prevents the negative feedback effect of estrogen, thus allowing FSH secretion to increase so that follicle development can be stimulated.
The aromatase inhibitors letrozole and anastrozole are approved for the treatment of breast cancer in post-menopausal women, but have been found to have some efficacy in ovulation induction. These drugs prevent the conversion of androgens to estrogens. Just as clomiphene does, aromatase inhibitors work to induce ovulation by limiting estrogen’s negative feedback inhibition of gonadotropin secretion. The difference is that they do so by preventing estrogen formation, rather than antagonizing the estrogen receptor.
A potential advantage of aromatase inhibitors over clomiphene
is that they have a shorter half-life, allowing normal estrogen
action later in the cycle. In the mid-follicular phase, negative
feedback from estrogen limits gonadotropin secretion so that
only a single follicle becomes dominant. Thus, ovulation
induction with aromatase inhibitors should have less risk than
clomiphene of multiple ovulations. As well, the shorter
half-life allows more estrogen stimulation of endometrial
development during the proliferative phase in the uterus.
More studies are needed to compare aromatase inhibitors with
clomiphene; so far they have mainly been tried in patients who
have failed to respond to clomiphene.
This treatment approach addresses the problem of insulin resistance. Studies have shown that early interventions to improve insulin sensitivity can help prevent or slow the development of type 2 diabetes mellitus. Interestingly, increasing insulin sensitivity also seems to help improve ovarian function. Weight loss improves insulin sensitivity, and can also restore normal ovulatory cycles in some women with PCOS. Metformin is an insulin sensitizing drug that is used to treat PCOS, and studies show that in some women it is apparently safe and effective in lowering insulin and androgen levels, and increasing ovulation.
In some women, clomiphene is not successful at inducing ovulation. In this case, exogenous FSH is needed to stimulate follicle development. The first treatment developed was menotropin, a mixture of gonadotropins purified from the urine of menopausal women. (Can you think why this would be a particularly rich source of FSH and LH?) Although menotropin contains LH, it is really the FSH that is important for stimulating ovulation in women with PCOS. Urofollitropin is FSH purified from menopausal urine. More recently, purified recombinant FSH (follitropin) has been produced.
Treatment with gonadotropins is more expensive, and involves more risk. One problem is that high levels of FSH may induce multiple ovulations and cause higher order pregnancies (i.e. twins or triplets) which are risky for the mother and the developing fetuses. Another problem is ovarian hyperstimulation syndrome, a dangerous condition that arises when the ovary is stimulated so that multiple follicles mature. During ovarian hyperstimulation syndrome there is an increase in vascular permeability that leads to edema, nausea, and abdominal pain. If severe, it can result in clotting abnormalities, respiratory distress, and renal failure. Because of the risk of ovarian hyperstimulation, women treated with gonadotropins need to be carefully monitored for excessive increases in estrogen secretion.
Ovarian surgery is another approach that can be used in women
for whom clomiphene treatment is unsuccessful. There are
different techniques, but they all involve inducing damage to
ovarian tissue. The hypothesis is that a small amount of damage
to the ovary works to break the cycle of excessive androgen
production and abnormal negative feedback. While there are risks
involved with surgery, an advantage is that there is no risk of
ovarian hyperstimulation syndrome or multiple pregnancy.