Testosterone-induced increase of insulin-like growth factor I levels depends upon normal levels of growth hormone Giuseppe Saggese, Graziano Cesaretti, Giulia Franchi and Luisa Startari Endocrine Unit, Department of Pediatrics, University of Pisa, Pisa. Italy Saggese G, Cesaretti G, Franchi G, Startari L. Testosterone-induced increase of insulin-like growth factor I levels depends upon normal levels of growth hormone. Eur J Endocrinol 1996;135:211\p=n-\5. ISSN 0804\p=n-\4643 Pubertal development is associated with a rise in plasma insulin-like growth factor I (IGF-I) levels that is related both to the increase in sex steroids and/or to the sex steroid-induced augmentation in endogenous growth hormone (GH) secretion. In order to investigate the relationship between IGF-I. GH and testosterone, we examined 42 male subjects with various clinical conditions (classical GH deficiency (CGHD, N = 5), non-classical GH deficiency (NCGHD, N = 7), short idiopathic stature (N = 6), nutritional obesity (N = 8), GH-treated CGHD (N = 4), GH-treated NCGHD (N = 5) and normal stature (N = 7)) in which, for evaluation of hypogonadism (i.e. the absence of one or both testes from the scrotal sac), human chorionic gonadotropin (hCG) tests were performed. We measured IGF-I, total and free testosterone and dehydroepiandrosterone sulfate (DHEAS) by radioimmunoassays before and 48 and 96 h after the start of the test. The values of IGF-I were lower (0.001 < p < 0.005) in CGHD and NCGHD than in the other groups. In comparison to basal levels, IGF-I values increased (0.005 < p < 0.05) both 48 and 96 h after the start of the hCG test in short idiopathic and normal stature children and in GH-treated subjects with NCGHD, but only 96 h in subjects with untreated NCGHD and GH-treated CGHD. No difference was demonstrated in basal values of total testosterone among any of the groups, while basal free testosterone levels were higher (0.001 < p < 0.05) in GH\x=req-\ treated subjects with NCGHD than in all the other groups except nutritional obesity; furthermore, free testosterone was higher (p < 0.05) in nutritional obesity than in CGHD. The values of total and free testosterone obtained both 48 and 96 h after the start of the hCG test were higher (0.001 < p < 0.05) than basal values in all groups. The DHEAS values did not show any significant change during the hCG test. Basal values were higher (0.01 <p < 0.05) in nutritional obesity than in the other groups. Considering all groups, chonological age, bone age and bone age/chronological age ratio were correlated with basal free testosterone, IGF-I and DHEAS levels (0.001 < p < 0.05), while basal free testosterone and IGF-I values were correlated with DHEAS levels (p < 0.005 and <0.01, respectively). In conclusion, our study during the hCG test in boys with various clinical conditions demonstrated an increase in IGF-I concentrations only in those boys with sufficient GH secretion or GH replacement therapy. These findings indicate that both sex steroids and GH are necessary to allow for the pubertal increase in IGF-I levels. Giuseppe Saggese, Endocrine Unit, Department of Pediatrics, University of Pisa, Via Rome 67, 56125 Pisa, Italy Pubertal development is associated with a rise in plasma IGF-I levels that is principally related to the increase in sex steroids (1-12): in male subjects a significant correlation with testosterone concentrations was also reported (2, 3, 6, 9). Furthermore, subjects with central precocious puberty showed an augmentation in IGF-I concentrations similar to that occurring at the onset of pubertal development in normal children (3.4, 13-15), and its reduction when sex steroids output was suppressed (3,4); on the other hand, in delayed puberty the increase in IGF-I occurred later than in normal subjects (2, 14). In all these reported subjects, GH secretion was normal with a physiological increase during puberty, regardless of the moment of onset of pubertal development. Therefore, the pubertal increase in IGF-I may be due also to the sex steroid-induced augmentation in endogenous GH secretion. Recently, Cara et al. (15) did not report an increase in IGF-I levels with the onset of precocious puberty in some GH-deficient children. The administration of human chorionic gonadotro¬ pins (hCG) is widely used in subjects with an empty scrotal sac without palpable testes, in order to assess the existence of their testes, or in subjects with suspected hypogonadism. The hCG-induced stimulation of the testes simulates for a short period the physiological activation that occurs at the onset of puberty and, therefore, may be used to evaluate the influence of sexual steroids upon the GH-IGF-I axis. Downloaded from Bioscientifica.com at 11/19/2018 12:02:38PM via free access