To evaluate the effect of growth hormone on testicular response to human chorionic gonadotropins (hCG) in vivo in humans, we selected patients with combined deficits of GH and gonadotropins who were in substitution treatment with both GH (from the time of diagnosis) and gonadotropins (from the time of induction of puberty). Testicular response to gonadotropin therapy was then evaluated during and six months after the cessation of GH treatment. Blood samples were collected before and 2, 4 and 6 days after hCG administration. hCG responses were calculated and expressed as the areas under the response curve. We studied four hypogonadotropic patients (aged 18-19 years) with associated GH deficiency. Their gonadotropin treatment consisted of hCG 1500 IU every six days, and FSH 75 IU every three days. The GH therapy replacement consisted of 4 IU thrice weekly. Testosterone, androstenedione, 17alpha-hydroxyprogesterone and estradiol were measured. In all subjects the testosterone area during GH treatment was significantly higher compared to the testosterone area obtained without GH administration (2993 +/- 1091 vs 2 310 +/- 7 5 1; M +/- SD; p < 0.04). The androstenedione area followed a similar pattern (708 +/- 3 77 vs 402 +/- 248; M +/- SD; p < 0.05). The 17alpha-hydroxyprogesterone area, on the contrary, was significantly higher during GH withdrawal (542 +/- 30 7 vs 2 3 5 +/- 190; M +/- SD; p < 0.05). As far as the estradiol area is concerned, no significant differences were found (22860 +/- 10082 vs 25697 +/- 13640; M +/- SD). In Conclusion, GH administration seems to improve testosterone production induced by human chorionic gonadotropins. The finding of the inverse response pattern of 17alpha-hydroxyprogesterone with respect to testosterone led us to suppose that the increased testosterone area during GH treatment may be due to an increased activity of C17,20-lyase enzyme.