Characterization of functionally typical and atypical types of polycystic ovary syndrome

Abstract

Context: The typical polycystic ovary syndrome (PCOS) phenotype includes 17-hydroxyprogesterone (17OHP) hyperresponsiveness to GnRH agonist (GnRHag) testing. Functionally atypical PCOS lacks this feature.

Objective: The hypothesis was tested that the typical PCOS ovarian dysfunction results from intrinsically increased sensitivity to LH/human chorionic gonadotropin (hCG) due to a flaw in FSH action. PARTICIPANTS/DESIGN/INTERVENTIONS/MAIN OUTCOME MEASURES: After phenotyping a cohort of 60 women, steroid and inhibin-B responses to gonadotropins were evaluated in representative typical (n = 7) and atypical (n = 5) PCOS and healthy controls (n = 8). Submaximal hCG testing before and after an FSH test dose was performed in random order before and after prolonged ovarian suppression by depot GnRHag.

Setting: The study was performed at a Clinical Research Center.

Results: Of our PCOS cohort, 68% were the typical type. Typical PCOS had 17OHP hyperresponsiveness and, unlike controls, significant androgen and estradiol responses to hCG. FSH increased inhibin-B and did not inhibit free testosterone or enhance estradiol responsiveness to hCG, all unlike controls. After ovarian suppression, 17OHP, androstenedione, and inhibin-B responsiveness to gonadotropin testing persisted. Atypical PCOS had significantly higher body mass index but lower ovarian volume and plasma free testosterone than typical PCOS. Steroid responses to hCG were insignificant and similar to controls. FSH suppressed free testosterone but stimulated inhibin-B. The estradiol level after combined hCG-FSH was subnormal. Free testosterone was less GnRHag suppressible than in typical PCOS.

Conclusions: Typical PCOS is characterized by intrinsic ovarian hypersensitivity to hCG to which excessive paracrine FSH signaling via inhibin-B may contribute. Atypical PCOS is due to a unique type of ovarian dysfunction that is relatively gonadotropin hyposensitive.

Figure 1

Simplified version of the two-gonadotropin, two-cell model of the normal regulation of ovarian steroidogenesis during the early to midfollicular phase of the menstrual cycle. This diagram shows the major relationships evaluated in this study. LH (or the LH analog hCG) stimulates theca cells to form androgen. Androgens are used as substrates for estrogen formation by granulosa cells, a process requiring aromatase activity, which is normally up-regulated by FSH in accord with the state of maturation of granulosa cells. Gonadotrope secretion of FSH is more sensitive than LH to endocrine negative-feedback regulation by the granulosa cell products estradiol and inhibin (fine dotted line). FSH-dependent granulosa cell factors, including estradiol and inhibin-B, also act in an intraovarian paracrine manner to modulate the androgenic response to LH/hCG positively (+) or negatively (−) in accord with granulosa cell needs (heavy dotted lines). A’dione, Androstenedione; T, testosterone. Based on model of Ehrmann et al. (8).

Figure 2

Steroid responses to gonadotropin sensitivity tests. Steroidogenic responses (mean ± sem) to hCG, FSH, and hCG plus FSH before and after 3–4 months of leuprolide depot administration in typical (TYP) PCOS, atypical (ATYP) PCOS, and healthy volunteers (CON). The mean of basal samples obtained before administering hCG and before administering FSH is displayed. The number of subjects in each group before (n = 5–8) and after (n = 5) depot leuprolide treatment is indicated in the 17OHP panels.

Figure 3

Inhibin-B responses to gonadotropin sensitivity testing. Subjects and study results are displayed as in Fig. 2.

Figure 4

Scatterplots showing changing relationships of inhibin-B to sex steroid levels in response to hCG, FSH, and their combination. Groups are designated as in Fig. 2. A’dione, Androstenedione; Testo, testosterone. Significant relationships are indicated by heavy linear regression lines; trend line is indicated by light dotted line. Note lower scales for inhibin on hCG and 17OHP on FSH. The correlations were virtually identical for total and free testosterone.