DISSOCIATION OF ADRENAL ANDROGEN AND CORTISOL SECRETION IN CUSHINGS-SYNDROME

OBJECTIVES While ACTH may modulate adrenal androgen production, there is evidence that other factors are required. Cushing's disease and ectopic ACTH secretion provide a little utilized opportunity to examine adrenal androgen levels in conditions of ACTH excess. We have compared plasma cortisol values with plasma levels of androstenedione, dehydroepiandrosterone (DHEA), DHEA-sulphate (DHEAS), testosterone and an index of free testosterone, the testosterone/sex hormone binding globulin ratio, prior to treatment in patients with Cushing's syndrome. PATIENTS AND MEASUREMENTS Plasma from 15 adult patients with Cushing's disease and three adults with the ectopic ACTH syndrome was obtained prior to treatment and submitted to specific immunoassays for the measurement of the above steroids. RESULTS Plasma cortisol values of 15 patients with Cushing's disease (range 326-1140 nmol/l, normal range 190-690 nmol/l were elevated in 9; in contrast, plasma androstenedione (4.1-11.3 nmol/l, normal range, men 2.1-7.7, women 3.3-9.9 nmol/l) was elevated in only two patients, plasma DHEAS (3.3-17.8 mu mol/l), normal range, men 4.5-18.4, women 3.5-11.8 mu mol/l) was elevated in only 4 patients and plasma DHEA (4.8-45.2 nmol/l, normal range 11-48 nmol/l) was normal or low in all 15 patients. Plasma androstenedione was markedly elevated (74 nmol/l) in one of three patients with ectopic ACTH syndrome, moderately elevated in another, and normal in the third patient. In contrast, plasma DHEA and DHEAS levels were suppressed in the patient with the highest androstenedione level and low or normal in the other two patients. CONCLUSIONS These data suggest that ACTH alone does not control adrenal androgen secretion. The data also suggest that variability in the processing of proopiomelanocortin (the precursor of ACTH and related peptides) occurring in Cushing's disease and ectopic ACTH syndrome may account for differences in the relation of cortisol to androgens observed between the disorders and when compared to that in normal subjects.