Although previous investigations have examined the importance of andro
gens in the regulation of the human menstrual cycle, no consensus has
been reached, due to conflicting results. We have therefore used the n
on-steroidal anti-androgen flutamide as a pharmacological probe to eva
luate the role of androgens in the control of gonadotropin secretion i
n normally cycling women. Eight women were studied during control and
treatment cycles, during which either placebo (as control) or flutamid
e (750 mg orally) was given daily. Blood was sampled every other day d
uring the follicular and luteal phases and daily around the expected m
idcycles for determination of luteinizing hormone (LH), follicle stimu
lating hormone (FSH), estradiol, progesterone and androgens (testo-ste
rone, androstenedione, dehydroepiandrosterone sulfate) by radioimmunia
ssay and immunoradiometric assay. To establish unstimulated and gonado
tropin releasing hormone (GnRH)-stimulated gonadotropin profiles, bloo
d samples were frequently collected (every 10 min for 8 h, GnRH 25 mug
i.v. after 7 h on day 10 in both the control and treatment cycles. Co
mpared to control conditions, the durations of both the follicular and
luteal phases did not change considerably during flutamide treatments
. Serum androgen levels (testosterone, androstenedione dehydroepiandro
sterone sulfate) were significantly (p < 0.01) reduced during androgen
antagonism. Daily gonadotropin and estradiol levels did not differ be
tween control and flutamide cycles, while progesterone secretion tende
d to be attenuated (p = 0.2) during the luteal phases of the flutamide
cycles. The LH and FSH secretory profiles and the GnRH-stimulated gon
adotropin responses remained virtually unchanged during androgen antag
onism. Thus, while androgen receptor blockade was capable of markedly
reducing the circulating androgen concentrations, it failed to modify
the temporal organization and endocrine characteristics of the menstru
al cycles significantly. In particular, the dynamics of unstimulated e
pisodic and GnRH-stimulated gonadotropin release remained unaffected.
Within the limitations of the nature of a pharmacological experiment,
we conclude that androgens may not critically participate in the neuro
endocrine regulation of the human menstrual cycle.