alone at a dose of 1 mg/kg was associated with the induction of a secretory endometrium, but higher doses (5 mg/kg) inhibited both proliferation and secretory activity. Thus, in the absence of progesterone, the actions of mifepristone were biphasic in nature. This mifepristone action appears to be associated with an increase in the estrogen-receptor content of the relevant endometrium, a dose-dependent phenomenon (Neulen et al., 1990). However, the latter observation may be affected to some degree by the choice of the denominator—that is, protein versus cellular DNA content.

More recently, the effects of mifepristone were examined using rats in which endometriosis had been experimentally induced (Tjaden et al., 1993). Animals were treated daily with mifepristone (10 mg/kg body weight subcutaneously) for two, four, six, or eight weeks. No regression of endometriosis was observed. The precise reasons for this apparent species/model specificity remain unknown.

Stimulated by the preceding observations, Kettel and associates (1991) reported the first relevant (though preliminary) human study on the endocrine response to long-term administration of mifepristone to patients with endometriosis. Six women who had normal menstrual cycles and who had endometriosis were recruited for the study, and received mifepristone at a dose of 100 mg/day for three months. Importantly, this regimen produced uniform amenorrhea. The mean circulating levels of luteinizing hormone (LH) and the LH pulse amplitude (but not frequency) were increased. An antiglucocorticoid effect was apparent as assessed by an increase in the circulating levels of cortisol and adrenocorticotropic hormone (ACTH). Pelvic pain was reportedly lessened in all subjects, but there was no significant change in the extent of the disease as evaluated by follow-up laparoscopy.

Unpublished observations reported by Yen at the Institute of Medicine workshop addressed the administration of mifepristone for a period of six months. Nine women with symptomatic endometriosis, who had no improvement with other forms of medical therapy, were recruited and treated with 50 mg/day mifepristone. Again, amenorrhea was uniformly induced. Likewise, all subjects reported a significant decrease in pelvic pain and dysmenorrhea. American Fertility Society (AFS) scores, a measure of the extent of disease, improved significantly in eight of nine subjects. Bone mineral density measurements of the lumbar spine and femur revealed no adverse effect. An initial increase in the circulating levels of LH and testosterone was noted in the first month of treatment. The levels of estradiol remained in midfollicular phase range. No antiglucocorticoid effect was noted. Side effects included transient, mild increases in liver transaminase (20 percent of patients), which returned to normal after one month. A follow-up study using a lower dose of 5 mg/day of mifepristone administered for six



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement