The effect of growth hormone replacement therapy on adrenal androgen secretion in adult onset hypopituitarism.

OBJECTIVE: Growth hormone replacement therapy in GH-deficient children is associated with enhanced adrenal androgen production, raising the possibility that GH might stimulate adrenocortical hormone secretion. This has not been extensively investigated in adults to date. GH is a potent modulator of the activity of the 11beta-hydroxysteroid dehydrogenase type 1 (11beta-HSD1) enzyme and by altering cortisol metabolism can affect the function of the hypothalamo-pituitary-adrenal (HPA) axis and therefore potentially of adrenal androgen secretion. This study examined the effects of GH replacement in GH-deficient adults on adrenal androgen secretion.
DESIGN: Prospective study of the effect of GH replacement therapy on adrenal androgen production in patients with adult onset hypopituitarism over a 12-month period. PATIENTS AND METHODS: Thirty adult GH-deficient patients were classified into two groups according to their cortisol responses to an insulin-induced hypoglycaemia or a glucagon stimulation test: 13 patients were adrenocorticotropic hormone (ACTH)-sufficient (nine females, age 45.1 +/- 3 years), whereas 17 patients were ACTH-deficient (11 females, age 45.5 +/- 3 years). Serum samples were collected before patients were initiated on GH replacement therapy using a dose titration regimen, and after 6 and 12 months on GH therapy for measurement of serum IGF-I, dehydroepiand-rosterone sulphate (DHEAS), Delta4-Androstenedione (A4), testosterone, cortisol, sex hormone binding globulin (SHBG) and cortisol binding globulin (CBG).
RESULTS: Six months after the initiation of GH replacement therapy, serum IGF-I levels were within the normal age-related reference range in both groups of patients and this was maintained at 12 months [in all patients 0 vs. 6 months: median (interquartile range): 92.5 ng/ml (73-116 ng/ml) vs. 191 ng/ml (159-224 ng/ml), P < 0.01]. In both ACTH-sufficient and -deficient groups of GH-deficient patients, pretreatment serum DHEAS levels were lower than the normal age-related reference range (P < 0.01); the ACTH-deficient patients had significantly lower DHEAS levels than the ACTH-sufficient patients [median (interquartile range): 0.5 micro mol/l (0.4-1.2 micro mol/l) vs. 1.5 micro mol/l (0.6-2.7 micro mol/l), P < 0.05]. Following GH replacement therapy, median levels of serum DHEAS levels rose from 1.5 micro mol/l (0.6-2.7 micro mol/l) to 1.9 micro mol/l (1.9-3.9 micro mol/l) in ACTH-sufficient patients, increasing in 11 of the 13 patients (P < 0.02). In this group, the median percentage increase from baseline was 32% at 6 months (P < 0.05). In contrast, baseline serum DHEAS levels [0.5 micro mol/l (0.4-1.2 micro mol/l)] declined in or from the measurable range in 47% of ACTH-deficient patients [median -16%; range -36-0] and only in one patient a + 0.2 micro mol/l increase was observed. GH dose requirements tended to be lower in ACTH-sufficient patients [1.2 U/day (0.8-1.4 U/day) vs. 1.6 U/day (1.0-2.0 U/day); P = 0.062]. There were no significant changes in serum testosterone, A4, SHBG and/or CBG levels, compared to the pretreatment levels, in either group of patients over the 12 months of GH replacement.
CONCLUSIONS: This study shows that median serum DHEAS levels are significantly lower in GH-deficient patients, even those with intact ACTH reserve, than in aged-matched controls. GH replacement therapy is associated with a significant increase in mean serum DHEAS only in ACTH-sufficient patients. These findings are consistent with either (i) GH stimulation of adrenal androgen production in the permissive presence of ACTH or (ii) an inhibitory effect of GH on 11beta-HSD type 1 activity leading to enhanced cortisol clearance, subsequent activation of the HPA axis and ACTH-mediated androgen secretion.