OBJECTIVE To test the hypothesis that adrenal steroidogenesis in nonclassic adrenal hyperplasia (NCAH) patients is, at least in part, independent of adrenocorticotropic hormone (ACTH) control. METHODS Prospective controlled clinical study. METHODS Patients and healthy volunteers in an academic research environment. METHODS Four patients with 21-hydroxylase (21-OH) deficient NCAH and four healthy control women. METHODS Patients received the long-acting gonadotropin-releasing hormone analog (GnRH-a) leuprolide acetate (3.75 mg/month IM) on weeks 0 and 4; and dexamethasone (DEX) in weekly incremented doses (0.25 mg/day, 0.50 mg/day, 1.0 mg/day, and 1.5 mg/day), beginning on weeks 4, 5, 6, and 7, respectively. METHODS The levels of 17-hydroxyprogesterone (17-HP), progesterone (P4), androstenedione (A4), dehydroepiandrosterone sulfate (DHS), and cortisol (F) were measured at the beginning of weeks 0, 4, 5, 6, 7, and at the end of the study (beginning of week 8). RESULTS Patients and controls had a similar median age and body mass index (BMI). There were no significant decreases in the median levels of the studied hormones following 4 weeks of treatment with GnRH-a only, in either NCAH patients or controls. Analysis of individual hormonal values demonstrated that by the end of the study (after DEX of 1.5 mg/day during a week) only 2 of 4, 0 of 4, 3 of 4 and 3 of 4 NCAH patients had 17-HP, P4, A4, and DHS levels within the range of control values, respectively. CONCLUSIONS Ovarian and incremental adrenal suppression did not fully suppress progestogen and androgen production in all of the study patients with 21-OH-deficient NCAH, suggesting that their production was partially independent of ACTH stimulation. Potentially in these patients subtle degrees of adrenocortical hyperplasia and/or abnormal enzymatic kinetics are responsible for the nonsupressibility.