Healthy postmenopausal women aged 60-65 years (n=150) were randomized to receive twice daily application on the skin of 3g of a 0.3% dehydroepiandrosterone (DHEA) or placebo emulsion for 12 months. Serum DHEA and eleven of its metabolites were measured at screening and on day 1, as well as at 1, 3, 6, 9 and 12 months to study long-term metabolism. While serum DHEA and androst-5-ene-3beta, 17beta-diol (5-diol) increased by 203% and 178%, respectively, on average, during the 12-month period, the sum of concentrations of the metabolites of androgens, namely androsterone glucuronide (ADT-G), androstane-3alpha,17beta-diol-3G and -17G increased by only 71% while usually non statistically significant changes of 30%, 17% and 20% were observed for estrone (E(1)), estradiol (E(2)) and E(1) sulfate (E(1)-S), respectively. Despite the return of serum DHEA to normal premenopausal values with the present DHEA treatment regimen, the 65% decrease in the androgen pool found in this group of postmenopausal women is in fact corrected by only 24%, thus remaining 41% below the values found in normal premenopausal women. In fact, the changes in serum DHEA observed after percutaneous DHEA administration are a 186% overestimate of the true changes in androgen formation while the overestimate of estrogen production is even much higher. On the other hand, the pharmacokinetics of the steroids are stable over the 12-month period with no significant induction or decrease of activity of the enzymatic systems transforming DHEA predominantly into androgens.