represent an ideal treatment of Cushing's syndrome. The short-term studies discussed above demonstrated that the compensatory homeostatic mechanisms of ACTH-producing tumors would probably preclude effective therapy, because the tumor would likely continue to increase ACTH production. In patients with long-standing hypercortisolism not caused by Cushing's disease, however, RU 486-induced antagonism of peripheral glucocorticoid effects presumably would not be overcome. This concept was tested in a few patients with ectopic ACTH secretion and proved correct. At daily doses of 5–22 mg/kg, RU 486 reversed psychosis, hypokalemia, hypertension, weight gain, inhibition of luteinizing hormone, follicle-stimulating hormone, testosterone, thyroid-binding globulin (TBG), corticosteroid-binding protein (CBG), and T4, and restored euglycemia. In most patients there was little effect on cortisol or ACTH levels, although cortisol, but not ACTH levels, fell in an occasional patient, suggesting a potential effect on steroidogenesis (Nieman et al., 1985; Chrousos et al., 1989). The compound worked best in patients with sustained, invariant hypercortisolism. We have given RU 486 to seven such patients for six weeks to a year, with excellent results, either as preoperative preparation or while seeking to localize and remove an ACTH-secreting tumor. Three patients had no adverse side effects. Although no hepatic, hematologic, renal, or dermatologic toxicity was observed, three patients had nausea, one with prostration reminiscent of adrenal insufficiency. Two of three men developed gynecomastia (presumably because of antiandrogenic properties), and one had an unmasking of autoimmune thyroid disease (not uncommonly seen after successful reversal of Cushing's syndrome). RU 486 was discontinued in one patient with variable cortisol levels, for whom it was not possible to find an optimal daily dose. The agent was discontinued within three weeks, and dexamethasone was instituted, in three others in whom hypotension and clinical deterioration suggested either sepsis or RU 486-induced adrenal insufficiency. Of these, one developed Pneumocystis carinii pneumonia. Thus, while RU 486 can be an effective agent for the treatment of hypercortisolism, it is difficult to monitor therapy and adjust dose because of the temporal pattern of response, and the risk of adrenal crisis alternating with undertreatment is great in patients with variable hormonogenesis.

RU 486 has potential activity as a tumoricidal agent in tumors with glucocorticoid or progesterone receptors, and has been tested in breast cancer, leiomyomas (reviewed elsewhere in this report) and meningiomas. The effects of RU 486, 200 mg/day, given for up to a year, have been reported in 38 patients with meningioma (Lamberts et al., 1991; 1992; Grunberg, 1993). Regression of tumor size was documented by objective measures in about 30 percent of patients; in the remainder, responses were evenly divided between tumor stabilization and growth.