Elevated serum levels of free insulin-like growth factor I in polycystic ovary syndrome

Polycystic ovary syndrome (PCOS) is the most common cause of anovulation in women. Previous studies suggest that the pathogenesis of PCOS may involve interrelated abnormalities of the insulin-like growth factor (IGF) and ovarian steroidogenesis systems. We investigated this hypothesis in fasting serum samples from 140 women with PCOS (age, 27.4 ± 0.4 yr; body mass index, 26.3 ± 0.5 kg/m²; mean ± SEM). IGF-related parameters were also studied in a group of normoovulatory women (n = 26; age, 26 ± 4 yr; body mass index, 23.6 ± 4.3 kg/m²). For the PCOS group, the mean testosterone (T) level was 2.5 ± 0.1 nmol/L, and it was significantly correlated with LH (r = 0.41; P < 10^-6), estrone (r = 0.33; P = 0.016), estradiol (r = 0.18; P = 0.04), and androstenedione (AD; P < 10^-6), but not with dehydroepiandrosterone sulfate (P = 0.71), a marker of adrenal steroidogenesis. T and AD were also related to total ovarian follicle number and ovarian size, as previously found with normoovulatory women (1). There were no differences between the PCOS subjects and the normoovulatory group for total IGF-I, IGF-II, or IGF-binding protein-3 (IGFBP-3). However, IGFBP-1 levels were significantly decreased in the PCOS group (1.0 ± 0.2 vs. 7.3 ± 1.1 ng/mL; P < 0.001) and were inversely correlated with serum insulin levels (r = -0.50; P < 10^-8). Serum levels of free IGF-I (fIGF-I) were elevated (5.9 ± 0.3 vs. 2.7 ± 0.3 ng/mL; P < 0.001) in inverse relation with IGFBP-1 (r = -0.31; P = 0.046). Serum fIGF-I levels were related to total follicle number (r = -0.35; P < 10^-4) and to the ratio of sex hormone-binding globulin to T (r = -0.23; P = 0.009). However, these relationships were not independent of other variables. Despite the more than 2-fold elevation in fIGF-I levels, significant relationships between fIGF-I and markers of ovarian steroidogenesis (T, AD, estradiol, and estrone) could not be demonstrated. In conclusion, although we confirmed correlations between LH and hyperandrogenemia and have found abnormalities in the IGF system in a large cohort of PCOS subjects, a direct relationship between hyperandrogenism and the IGF system could not be shown. Previous studies suggest that elevated LH and hyperinsulinemia lead to excess ovarian androgen synthesis in PCOS and that the intraovarian IGF system is important for normal follicle development and may be important in the arrested state of follicle development in PCOS. However, the data presented in this cross-sectional study suggest that insulin-related changes in circulating IGFBP-1 and subsequent elevation of fIGF-I reflect insulin resistance and have little enhancing effects on ovarian steroidogenesis in this disorder.