Effects of metformin on serum insulin levels and ovarian steroidogenesis in nonobese women with polycystic ovary syndrome

ABSTRACT

Evidence indicates that nonobese women with polycystic ovary syndrome [PCOS] are insulin resistant and hyperinsulinemic. This work was conducted to study the hypothesis that hyperinsulinemia stimulates the steroidogenic enzyme P450c17 alpha activity in these patients, with consequent increase in serum androgen levels. We assessed the activity of the cytochrome P450c17 alpha by measuring the response of 17 alpha-hydroxypro- gesterone [17-PROG] to a GnRH agonist [leuprolide] in 30 normal-weight women with PCOS. Fasting and the 2h serum glucose and insulin, and serum levels of sex hormone binding globulin [SHBG], testosterone [TSN], free testosterone [F-TSN], dehydroepiandrosterone sulphate [DHEAS] estradiol [E[2]], progesterone [PGN], 17-PROG, LH, and FSH were measured before and after oral administration of either metformin 500 mg tab [in 18 patients] or a placebo [in 12 patients] t.d.s. for 8 weeks. We also estimated fasting and the 2h serum glucose and insulin, serum SHBG, and sex hormones in 12 normal controls before and after metformin administration. Serum glucose was measured by a hexokinase method and serum insulin was assayed by a double antibody technique. 17-PROG and DHEAS concentrations were determined by radioimmunoassay after chromatographic purification. F-TSN and SHBG were measured by a competitive protein binding assay. An electrochemiluminescence immunoassay intended for use on Roche Elecsys 1010/2010 analyzers was used for determination of, TSN, E[2], PGN, LH and FSH. The metformin and placebo groups of PCOS women did not differ in both the fasting and the 2h serum glucose values whether before or after treatment. However, these values in the control group showed a significant decrease [p<0.001] as compared to the metformin group. Metformin administration was associated with significant [p<0.01] decrease both in the fasting and the 2h serum insulin values in the metformin group and the control women but not in the placebo group. Serum SHBG was significantly increased on metformin therapy in PCOS women [p<0.001] but not in the placebo group nor in the controls. The metformin group showed a significant posttreatment decrease in serum TSN, F-TSN, E[2] and DHEAS and a significant increase in Serum PGN both before and after leuprolide administration [p<0.001]. There was a nonsignificant change in serum sex steroids, on treatment, in the placebo group. In the metformin group, the mean basal and peak levels of 17-PROG decreased significantly [p<0.001] after treatment. As for serum LH, the metformin group showed a significant decrease in the mean basal levels [p < 0.05], the early [p < 0.05] and late serum responses [p < 0.001] to leuprolide after metformin treatment. These values for LH and 17-PROG did not change [on treatment] in the placebo group. These findings suggest that Hyperinsulinemia stimulates ovarian cytochrome P450c17 alpha activity in nonobese women with PCOS. They also indicate that decreasing serum insulin with metformin reduces cytochrome P450c 17 alpha activity and ameliorates the hyperandrogenism of these women. Improvement of the endocrine profile of PCOS on metformin therapy was also discussed