Serum total prostate-specific antigen assay in women with Cushing's disease or alcohol-dependent pseudo-Cushing's...

The distinction between Cushing's disease (Cushing's syndrome dependent on adrenocorticotropic hormone (ACTH)-secreting tumors of pituitary origin) and pseudo-Cushing's states (Cushingoid features and hypercortisolism sometimes present in alcoholic, depressed or obese subjects) can present a diagnostic challenge in clinical endocrinology. Recently, the availability of a highly sensitive immunofluorometric assay for the measurement of total prostate-specific antigen (PSA) provided the possibility to measure serum PSA levels in women. Interestingly, PSA gene expression and protein production has been found to be upregulated by steroid hormones, such as androgens, glucocorticoids, mineral corticoids and progestins. In fact, serum total PSA concentrations appear to be higher in female patients with Cushing's disease than in normal women. We wondered whether a similar phenomenon also occurs in pseudo-Cushing's state.

In order to answer this question, we compared the serum total PSA levels measured in 10 female subjects with alcohol-dependent pseudo-Cushing's state with those observed in 8 female patients with Cushing's disease and in 15 age-matched healthy women. Serum testosterone, ACTH and cortisol, and 24-hour urinary cortisol levels were measured; cortisol suppression after dexamethasone was also tested in all subjects.

The basal serum levels of ACTH and cortisol were significantly lower in normal subjects than in patients with Cushing's disease or pseudo-Cushing's state; these latter groups showed similar basal hormonal values. Dexamethasone administration was unable to suppress serum cortisol levels in 5 subjects with Cushing's disease and 6 subjects with pseudo-Cushing's state. Serum testosterone values in the group with Cushing's disease were higher than in the other groups. No differences were observed between pseudo-Cushing's and normal subjects. Serum total PSA levels were significantly higher in women with Cushing's disease than in subjects with pseudo-Cushing's state and normal controls; these latter groups showed similar PSA values. When serum total PSA and testosterone levels were considered together, a significant positive correlation was observed in the group with Cushing's disease, but not in the other groups.

These data indicate that the steroid milieu responsible for the elevation in serum PSA in women with Cushing's disease is not present in subjects with alcohol-dependent pseudo-Cushing's state, suggesting the possible use of PSA as a marker of differentiation between these pathological conditions in women.