Combined low-dose pioglitazone, flutamide, and metformin for women with androgen excess.

CONTEXT AND

OBJECTIVE

One of the treatments for hyperinsulinemic hyperandrogenism in nonobese women is combined androgen receptor blockade (with flutamide; Flu), insulin sensitization (with metformin; Met) plus an estroprogestagen contraceptive. We tested whether adding low-dose pioglitazone (Pio; 7.5 mg/d) confers more benefit.

SETTING

The study was conducted at a university hospital.

STUDY POPULATION AND

DESIGN

This double-blind study enrolled 38 young women with hyperinsulinemic hyperandrogenism [mean body mass index (BMI) 24 kg/m(2)], all of whom started on Flu (62.5 mg/d) and Met (850 mg/d) plus a transdermal estroprogestagen, each for 21 of 28 d over 6 months. Patients were randomly assigned to receive, in addition, placebo (n=19) or Pio (n=19; 7.5 mg/d) for the same 21 of 28 d over 6 months.

MAIN OUTCOMES: BMI, waist to hip ratio, hirsutism score, fasting endocrine-metabolic markers, body composition, abdominal fat (visceral vs. sc), and carotid intima-media thickness were measured at study start and after 6 months.

RESULTS

PioFluMet reduced intima-media thickness more than FluMet and lowered glucose, IGF-I, and C-reactive protein more as well as the ratio of low-density lipoprotein to high-density lipoprotein cholesterol and the ratio of neutrophils to lymphocytes. PioFluMet treatment was followed by a leaner body composition and a loss of visceral fat (both P < 0.001). In the total group, the changes included not only decreases in waist to hip ratio, hirsutism score, and testosterone (all P < 0.001) but also minor drops in alanine aminotransferase, aspartate aminotransferase, gamma-glutamyl transpeptidase, and lactate dehydrogenase (all P < 0.005), indicating absence of hepatotoxicity; BMI remained unchanged. Clinical side effects were not detected.

CONCLUSION

In this proof-of-concept study, addition of Pio to FluMet plus an estroprogestagen led to improvements in the endocrine-metabolic condition, in low-grade inflammation, in total and visceral adiposity, and in markers of cardiovascular health.