Recent rat studies suggest that early exposure to exogenous testosterone accelerates the loss of androgen receptors and compromises eventual penile length. In humans we hypothesize that down regulation of the androgen receptor is not the mechanism that stops penile growth. To test this hypothesis we investigated the effects of androgen deprivation and supplementation on the developing human penis.

A total of 15 normal human fetal penises at 7 to 19 weeks of gestation (mean plus or minus standard deviation 12 −/− 4.5) was divided in half sagittally. Specimens were grafted beneath the renal capsule of male athymic nude mice or nude rats. Three groups of host animals were prepared, including 10 with no testosterone that were castrated at grafting, 15 with testosterone and 5 with super testosterone in which 50 mg. testosterone propionate pellets were implanted subcutaneously at grafting. Each fetal penile specimen was its own control, since half was implanted into an intact animal and the other into a castrated or super testosterone host. Six weeks after grafting the specimens were analyzed for gross size (length), histology and expression of androgen receptors.

All human fetal penile specimens grew from the nadir size and appeared as white exophytic growths on the surface of the host kidneys. Normal grafts were larger than castrate specimens (mean 6.9 −/− 2.1 versus 3.9 −/− 2.1 mm., p = 0.014). Mean length of the super testosterone specimens (7.3 −/− 2.3 mm.) was not significantly greater than that of normal specimens (p = 0.797). Histological analysis revealed that all specimens were composed of viable penile tissue. Cellular density of the castrate penises was approximately 2 times greater than that of the normal and super testosterone specimens (40.6 −/− 5.9 versus 25.1 −/− 2.8 cells per cm.^[2], p >0.001), as calculated on enlarged micrographs. Supraphysiological doses of testosterone did not change the histology compared to controls. Immunohistochemical localization revealed androgen receptors expressed throughout the corporeal bodies, surrounding stroma and penile skin with intracellular localization to nucleus. The mean proportion of cells expressing androgen receptors was higher in the castrate (29.4 −/− 5.2 cells per cm.^[2]) than in the normal (24.0 −/− 3.7) and super testosterone (24.7 −/− 4.5) grafts (p = 0.005). However, in regard to growth there was no change in the proportion of androgen receptor positive cells among the groups.

Testosterone influences penile growth, possibly as a result of extracellular stromal expansion. The number of androgen receptor positive cells in the human fetal penis did not change among the castrate, normal and super testosterone hosts. These experiments support the hypothesis that penile growth cessation is mediated by mechanisms other than down regulation of the androgen receptor. Furthermore, these data support the hypothesis that early administration of androgen to prepubertal male individuals does not result in a shorter phallus in adulthood.