CUTANEOUS APPLICATION OF AN ANDROSTENEDIONE GEL AS AN INVIVO TEST OF 5-ALPHA-REDUCTASE ACTIVITY IN WOMEN

Objective: Assessment of an in vivo test for 5alpha-reductase activity using serum markers, 5alpha-androstane-3alpha, 17beta-diol glucuronide and androsterone glucuronide, after the cutaneous application of androstenedione (A). Design: An A gel was applied for 6 days to the skin of normal women, male volunteers, and hirsute and nonhirsute patients with polycystic ovarian syndrome (PCOS). Blood samples were obtained at baseline and on day 6 of the A gel application. Blood samples were assayed for A, 5alpha-androstane-3alpha,17beta-diol glucuronide, and androsterone glucuronide. In three hirsute women, the protocol was followed before and after receiving an oral contraceptive (OC) and spironolactone 200 mg/d for 3 months. Setting: The study was performed in the outpatient clinic of the Division of Reproductive Endocrinology and Infertility, Women's Hospital of the Los Angeles County and University of Southern California Medical Center in Los Angeles, California. Patients, Participants: A total of eight nonhirsute patients with PCOS, seven hirsute patients with PCOS, and six male volunteers were enrolled in the study. Five normal women served as a control group. Main Outcome Measure: Serum A increased after 6 days by a similar magnitude in all groups. Serum androsterone glucuronide showed a significant increase from baseline only in the hirsute group (P < 0.03), whereas the increase in 5alpha-androstane-3alpha,17beta-diol glucuronide was not statistically significant. Results: The ratio of the increases in serum androsterone glucuronide over serum A was significantly higher in the hirsute group (P < 0.02). In the three hirsute patients who were placed on an OC and spironolactone, serum 5alpha-androstane-3alpha,17beta-diol glucuronide and androsterone glucuronide decreased after 3 months and did not increase with application of the gel for another 6 days. Conclusion: The cutaneous application of A provides a useful assessment of in vivo 5alpha-reductase activity. However, because we found that A absorption varied considerably (30% to 62%), we suggest that this in vivo test may not provide more information than baseline determinations of 5alpha-androstane-3alpha-17beta-diol glucuronide and androsterone glucuronide. It may, however, be useful as a parameter for assessing the effectiveness of various treatment regimens for hirsutism.