RESUMO
BACKGROUND: Serum androgen levels are below normal in patients with benign cortisol-secreting adrenocortical adenomas, owing to ACTH suppression. Associated androgen secretion is usually considered as indicative of malignancy. The objective of the study was to analyse the androgen-producing ability of cortisol-secreting adrenocortical adenomas. DESIGN: Retrospective data collection in a single referral hospital centre. METHODS: Dehydroepiandrosterone sulfate (DHEAS), Delta4androstenedione and testosterone concentrations were measured before and after adrenalectomy and then at 6-month intervals in 20 women (eight cortisol-secreting adrenocortical adenomas, six subclinical cortisol-secreting adrenocortical adenomas, and six nonfunctional adenomas). RESULTS: Before adrenalectomy, serum androgen concentrations were measurable in all women with clinically apparent and subclinical cortisol-secreting adrenocortical adenomas. DHEAS levels were either at the lower end of the normal range or below normal, but were always clearly detectable. Postoperatively, during adrenocortical insufficiency, DHEAS, Delta4androstenedione and testosterone concentrations fell to near the detection limit in all patients with cortisol-secreting adrenocortical adenomas (P = 0.008 for each marker) and showed a similar tendency to fall in all patients with subclinical cortisol-secreting adrenocortical adenomas. Pre- and post-treatment androgen concentrations did not differ in patients with nonfunctional adenomas. Immunohistochemical analysis confirmed CYP17, HSD3B2, SULT2A1 and CYB5 expression by all cortisol-producing tumours. The intensity of CYP17 and SULT2A1 expression was stronger in cortisol-secreting adenomas than in their adjacent normal adrenal tissue. CONCLUSION: Both clinically apparent and subclinical cortisol-secreting adrenocortical adenomas appear to show moderate autonomous androgen production. Thus, weak androgen secretion in patients with adrenocortical tumours should not necessarily be considered as a sign of malignancy.
