• endometrial cancer;
  • endometrial stromal sarcoma;
  • mifepristone;
  • RU-486;
  • hormone therapy


  1. Top of page
  2. Abstract
  6. Conflict of Interest Disclosures
  7. References


The objective of this study was to determine the efficacy of mifepristone (RU-486) in women with advanced or recurrent endometrioid adenocarcinoma or low-grade endometrial stromal sarcoma (LGESS).


Mifepristone (RU-486; 200 mg orally) was given daily to patients with progesterone receptor-positive advanced or recurrent endometrioid adenocarcinoma or LGESS. Patients were evaluated every 4 weeks for toxicity and response. Quality-of-life data were obtained using the Memorial Symptom Assessment Scale and Functional Assessment for Cancer Therapy.


Twelve of 13 enrolled patients were evaluable in the first phase of accrual. Stable disease was noted in 3 of 12 patients (at 8 weeks, 12 weeks, and ≥77 weeks, respectively), and the median time to disease progression was 48 days. Among the patients who had stable disease, 2 women had endometrioid endometrial cancer, and 1 woman had LGESS. No partial or complete responses were observed. The most frequent grade 1 and 2 toxicities were anorexia, fatigue, and mood alterations observed in 50%, 50%, and 58% of patients, respectively. The most common grade 3 toxicities were fatigue and dyspnea observed in 25% and 17% of patients, respectively. One patient experienced grade 4 dyspnea. Thirty-three percent of patients had asymptomatic elevations of corticotropin. No serious treatment-related adverse events occurred. There were no significant changes in quality of life.


Single-agent mifepristone used in the treatment of recurrent endometrioid adenocarcinoma or LGESS resulted in a stable disease rate of 25%. One patient who had a biopsy-positive disease recurrence remained stable at 77 weeks. Although mifepristone was tolerated well, as a single agent, it provided limited response as a single agent in women with progesterone receptor-positive uterine tumors. Recently, was been recognized that biologic agents used as single agents may result only in stable disease unless they are combined with cytotoxic agents. The authors concluded that further research into the best mode of application for mifepristone in the treatment of endometrial cancer is needed. Cancer 2009. © 2009 American Cancer Society.

Endometrial cancer is the most common cancer of the female genital tract in the US. It was expected to affect 40,100 women in the year 2008 and to cause 7470 deaths.1 Endometrioid histologies predominate and, of these, 50% to 60% are estrogen receptor (ER)/progesterone receptor (PR) positive.2, 3 Endometrial stromal sarcoma (ESS) is a rare tumor that comprises only 0.2% of all malignant uterine tumors. Low-grade ESS (LGESS) generally displays a slow growth pattern, and 37% to 60% of these tumors recur many years after resection of the primary tumor.4 The majority of LGESS are positive for ER, and many are positive for PR.3, 5-7

Therapy options for patients with recurrent/advanced endometrial cancer are limited. Treatment with radiation and chemotherapy has produced only limited success.8 However, the presence of ER and PR has been correlated directly with survival and response to hormone therapy and has been correlated inversely with tumor grade.9 Progesterones used for the primary treatment of well differentiated and recurrent endometrioid endometrial cancer have been associated with a response rates from 18% to 25% and stable disease rates from 20% to 50%.3, 10, 11 Used in the treatment of LGESS, progesterone-associated responses range from 33% to 45%, once again justifying the use of hormone-targeting agents in these tumors.12, 13 To our knowledge to date, the role of mifepristone (RU-486), a selective PR modulator (SPRM), in the clinical treatment of endometrial cancer has remained unexplored. It has been demonstrated that mifepristone has a dose-dependent in vitro inhibitory effect on endometrial cancer cell lines and can reduce the myometrial volume in benign leiomyoma.14-17 Although, the mechanism of inhibition is unknown, the regulation of PR concentration through vascular alterations of the endometrium and manipulation of p53 have been reported.15, 18, 19 This, this information served as a basis for the design of the current trial.

Mifepristone is a substituted 19-nortestosterone derivative that requires 2-step demethylation and hydroxylation to obtain its active form. It has a long half-life (18 hours), and there is no evidence of accumulation with chronic dosing.20 Mifepristone also exhibits antiglucocorticoid and weak antiandrogenic activity.20, 21 The antiprogestational activity of mifepristone results from competitive interaction with progesterone at PR sites. On the basis of studies in several animal species (mouse, rabbit, and monkey), the compound inhibits the activity of either endogenous or exogenous progesterone.22, 23 Mifepristone, as an antiprogestin, acts on the endometrium and blocks the action of progesterone at the cellular level by binding to the PR. The affinity of mifepristone for the PR is 5-fold greater than the endogenous progesterone. Consequently, it can produce a progesterone-like effect in the absence of progesterone.21, 24 Some have referred to mifepristone as a selective progesteron receptor modulator (SPRM), and this complex mode of action may be related to the interaction of mifepristone with the 2 isoforms of the PR (alpha and beta).25, 26

This SPRM action partially explains the growth-inhibitory effect of mifepristone on endometrial cancer cell lines.15, 16 Its primary abortifacient effect, however, is caused by the antiprogesterone and associated prostaglandin (PG)-inducing effects. It is believed that mifepristone sensitizes the uterus to PGs and, when administered in the first trimester of pregnancy at a single dose of 600 mg, causes pregnancy termination.22, 27, 28 It has been established that the effect of RU-486 on the endometrium is dependent on the timing of administration during the menstrual cycle.29 Since the design of this study, we have completed an in vitro investigation, and the results suggest that mifepristone also may work through an apoptotic effect to alter p53 and bcl-2.15 Others also have suggested that the combination of mifepristone with a PG synthesis inhibitor further increases the number of apoptotic, degenerative, and vascular changes induced by the drug in addition to the hormone mode of action noted in benign endometrium.27

Patients with recurrent uterine cancer, especially those burdened with high tumor volume and chemotherapy-resistant disease, often prefer less toxic regiments to aggressive chemotherapy, which has a minimal chance of producing long-term response rates. Patients prefer oral chemotherapy regimens to other forms of administration as long as efficacy is not sacrificed.30 The objective of this study was to evaluate the efficacy of mifepristone in patients with PR-positive, advanced or recurrent endometrial cancer or LGESS.


  1. Top of page
  2. Abstract
  6. Conflict of Interest Disclosures
  7. References

Women with histologically confirmed, inoperable, recurrent or metastatic endometrioid carcinoma of the endometrium or LGESS were eligible for enrollment. This study was conducted at The University of Texas M. D. Anderson Cancer Center after receiving approval from the Institutional Review Board for the protocol. Patients must have undergone previous radical surgery (minimum, total abdominal hysterectomy and bilateral salpingo-ophorectomy) and definitive radiotherapy or were not candidates for such procedures. Mifepristone was given orally at 200 mg once daily continuously (for a 4-week cycle). Mifepristone was obtained from Danco Laboratories, LLC (New York, NY). The decision to select this dose of mifepristone for this trial was based on safety data generated from the European clinical trials using mifepristone (200-400 mg daily) in women with recurrent or metastatic breast cancer and ovarian cancer.31-33

The trial required primary or recurrent tumors that were >10% PR positive. Positive hormone receptor status was defined as ≥10% PR-positive cells as determined by immunohistochemistry. Previous radiotherapy had to be completed at least 2 weeks before the initiation of mifepristone, and patients must have recovered from the acute side effects of such treatment. Women were required to have a Zubrod performance status ≤2 with an estimated life expectancy of at least 21 weeks. Hepatic, renal, or hematologic laboratory toxicities could be no greater than grade 1. Previous chemotherapy for recurrent or metastatic endometrial cancer was permitted. The patients must have had measurable disease as defined by the presence of bidimensionally measurable lesions with clearly defined margins on x-ray, scan (computed tomography [CT] or magnetic resonance imaging [MRI]), or physical examination.

Patients with a history of other malignancies were excluded. Patients also were excluded if they had uncontrolled hypercalcemia or if they were taking phenytoin, phenobarbital, or carbamazepine. Patients were excluded if they had a known predisposition to thromboembolic disorder or known unstable or untreated brain metastases. Women who were taking estrogen, progestin, antiestrogens, or other chemotherapy agents had to discontinue their use at least 3 weeks before beginning treatment with mifepristone.

Before starting mifepristone, patients were evaluated using CT or MRI scans, chest x-ray, and basic metabolic panel. Cortisol and corticotropin (ACTH) levels also were assessed at baseline and after the first 3 courses of treatment. If cortisol and ACTH results were normal, then no further testing was obtained unless the patient was symptomatic.

Quality-of-life and symptom data were obtained using the Functional Assessment of Cancer Therapy-General (FACT-G) and the Memorial Symptom Assessment Scale (MSAS).34, 35 Patients were asked to complete these surveys before they started treatment, after Cycles 1 and 2, and every third cycle thereafter until the end of the study. Response to treatment was evaluated at the end of Course 2 and every third course thereafter until discontinuation from the study.

Dose Modifications

Mifepristone was prescribed at 200 mg once daily. If patients developed grade 1 or 2 dermatologic side effects, then they stopped therapy until resolution of the rash. If the rash resolved within 14 days, then mifepristone was restarted at 100 mg once daily with the addition of oral diphenhydramine 25 mg once daily and oral cimetidine 300 mg 3 times daily. If the rash did not resolve to grade 0 or 1 or if the rash reappeared once it was treated, then the patient was removed from the study at the investigator's discretion. Patients with grade 3 dermatologic side effects were removed from the study.

Duration of Therapy

Mifepristone was given for at least 2 cycles (8 weeks) unless there was rapidly progressive disease. If the study had been continued, then patients who achieved a complete remission would have been continued on therapy for an additional 2 years after that remission, and patients who achieved either a partial remission or stable disease according to version 2.0 of the National Cancer Institute Common Terminology Criteria for Adverse Events (CTCAE) would have been continued on therapy until they developed disease progression.

Patients who experienced significant toxicities according to the CTCAE either had treatment interrupted until the toxicities resolved or were removed from the study at the discretion of the investigator. Patients who did not recover from treatment toxicities at least to grade 1 within 2 weeks of stopping the drug were taken off the study. Because of reports of Addisonian symptoms with the use of mifepristone, we tested for the development of Addisonian changes.20

Statistical Analysis

This was a single-arm, nonrandomized phase 2 trial of mifepristone for patients with PR-positive, advanced/recurrent endometrial endometrioid adenocarcinoma and LGESS. The study planned to enroll up to 37 patients using Simon's optimal 2-stage design to evaluate the first 12 patients in the first phase.36 The primary endpoint was overall response rate, including complete and partial responses. Stable disease was not considered a response. Activity was calculated as the proportion of patients with responsive disease (95% confidence interval [95% CI]). Treatment toxicity was evaluated. The Kaplan-Meier method was used to estimate time to progression and overall survival.

Sample Size and Stopping Rule

Simon's optimal 2-stage design was used.36 The null hypothesis was that the overall response rate would be <5% versus the alternative hypothesis that the overall response rate would be >20%. At the first stage, 12 evaluable patients were enrolled. Because no responses were observed, the trial was terminated.

This design yielded >.90 probability of a positive result if the true response rate was >20%. It yielded a >.90 probability of a negative result if the true response rate was <5% with >.54 probability of early negative stopping. Descriptive statistics were used to evaluate the quality-of-life data collected at specified time points. All P values <.05 were considered statistically significant.


  1. Top of page
  2. Abstract
  6. Conflict of Interest Disclosures
  7. References

Between November 2001 and February 2006, 13 patients were enrolled in the study. Patient and disease characteristics are outlined in Table 1. Two patients had LGESS, and 11 patients had endometrioid adenocarcinoma. Twelve patients were evaluable for response. One patient with endometrioid histology was not evaluable, because she had brain metastases identified immediately after enrollment. The mean patient age was 59 years (range, 29-77 years). Eight patients had received previous chemotherapy. The primary agents used were taxol and carboplatin; however, 1 participant each received either doxorubicin, etoposide, or gemcitabine in combination with platinum.

Table 1. Patient Demographics
CharacteristicNo. of Patients%
  1. LGESS indicates low-grade endometrial stromal sarcoma; ER, estrogen receptor; PR, progesterone receptor; XRT, external radiotherapy.

Median age (range), y59 (29-77) 
Histologic type  
Initial stage  
Receptor status  
 ER positive866.67
 PR positive12100
Site of recurrence  
 Outside of previous XRT field758.33
 Within previous XRT field325
 Both outside of and within216.67
Previous treatments  
 Previous chemotherapy866.67
 Previous hormone therapy650

All patients had developed recurrent disease. Seven patients (58%) developed recurrences strictly outside of a previous radiation field. Of these, 3 patients had received previous external pelvic radiation and brachytherapy. The sites of recurrence included the vaginal apex, liver, lung, periurethra, and sigmoid mesocolon and the aortocaval, para-aortic, and retroperitoneal lymph nodes.

Three patients (25%) developed disease recurrences strictly within a previous radiation field. Of these, all 3 had received previous external pelvic radiation, and 1 also had received radiation to the para-aortic lymph nodes. Two of patients had received previous brachytherapy. Recurrence sites in these patients included the iliac crest and sacrum, the common iliac lymph nodes, and retroperitoneal lymph nodes.

Two patients (17%) developed recurrences both within and outside of a previous radiation field. Both of these patients had received previous external pelvic radiation, and 1 patient had received previous brachytherapy. Sites of disease recurrence in these patients included the groin, vaginal cuff, pelvic and periaortic lymph nodes, liver, and lung.

Of the patients who had stable disease on mifepristone, 2 of 3 patients had received previous radiotherapy, and both developed recurrences within the radiated field. Furthermore, both of these women also had been exposed previously to taxol and carboplatin. Because of the small number of women with stable disease, it was difficult to determine whether the response was related to the biology of recurrences within a radiated field or chemoresistance.

All patients had confirmed PR-positive disease, and 8 patients (66%) had ER-positive disease. Three patients not evaluated for ER status. No patients died during the study. No patients had a clinical partial or complete response. Three patients had evidence of stable disease for 8 weeks, 12 weeks, and ≥77 weeks, respecitively. Nine patients had disease progression at the first evaluation time point (8 weeks). The median overall survival from the time of enrollment was 308 days (approximately 10 months), and the median progression-free survival was 48 days (Fig. 1).

thumbnail image

Figure 1. This chart illustrates the overall survival of evaluable patients.

Download figure to PowerPoint

Mifepristone was tolerated well: There were no dose reductions, and there were no delays because of drug toxicity. Greater than 30 cycles of treatment were completed. No serious treatment-related adverse events occurred. The most frequently reported grade 1/2 toxicities were anorexia, fatigue, and mood alterations (50%, 50%, and 58%, respectively). Ten patients had asymptomatic grade 1 increases in ACTH (83%), and 10 patients had asymptomatic grade 1 increases in serum cortisol (83%). The most common grade 3 toxicities were fatigue and dyspnea (25% and 17%, respectively). Only 1 patient experienced grade 4 dyspnea. A summary of adverse events is presented in Table 2.

Table 2. Common Toxicities (N=12)*
 Grade 1Grade 2Grade 3Grade 4
  • ACTH indicates corticotropin.

  • *

    Toxicities were graded according to version 2.0 of the National Cancer Institute Common Terminology Criteria for Adverse Events.

ACTH elevation1083000000
Cortisol elevation1083000000
Mood alteration5172171800

Eight serious adverse events were reported. All required hospitalizations. None were definitely drug related and, instead, appeared to be tumor related. One patient developed a superficial thrombosis that improved with ibuprofen and leg elevation. There was no evidence of deep vein thrombosis.

Twelve of 13 enrolled patients completed the baseline quality-of-life and symptom surveys, 7 patients completed surveys after Cycle 1, and 5 patients completed surveys after Cycle 2. There was an overall improvement in general quality-of-life scores from baseline until after Cycles 1 and 2, but these changes were not statistically significant. This pattern also was observed for the physical and functional subscales of the FACT-G. Similarly, there was an overall improvement in symptom assessment scores over time; however, again, these changes were not statistically significant. Specific subscale scores suggested that physical and psychological symptoms improved over time. Improvements also were noted for the Global Distress Index and for total MSAS scores.


  1. Top of page
  2. Abstract
  6. Conflict of Interest Disclosures
  7. References

In this population, treatment with mifepristone overall resulted in a 25% stable disease rate that lasted from 8 weeks up to ≥77 weeks at the time of last follow‒up. Seventy-five percent of patients experienced disease progression. These are disappointing results compared with the historic response rates reported with progesterone and megestrol acetate.9 This may be related to the lack of grade 1 tumors in the patients who enrolled in this trial. This also is unexpected given in vitro observations that mifepristone has a direct inhibitory effect on endometrial cell growth, specifically in the EM42, KLE, RL95-2, HEC-1-A, and Ishikawa cell lines, as well as apoptosis-inducing activity.15, 16, 37

Unfortunately, to our knowledge, no treatment for recurrent uterine cancer has been reported to prolong success rates significantly. A recent meta-analysis of 2471 patients who received various treatments, including progestogens, tamoxifen, aromatase inhibitors, gonadotropin releasing hormone analogs, and progestogen or progetogen/tamoxifen combination regimens for grade 1 through 3 endometrioid endometrial cancer, had the following response rates. Patients with grade 1 and 2 tumors had response rates that ranged from 14% to 100% with higher response rates in women who had PR-positive tumors9; and the Gynecologic Oncology Group studied alternating megestrol acetate/tamoxifen and reported a response rate of 38% in women with grade 1 tumors, 24% in women with grade 2 tumors, and 22% in women with grade 3 tumors.39

A growth-suppressive effect of mifepristone has been demonstrated in breast cancer cell lines in which both progesterones and antiprogesterones inhibited the growth of PR-positive carcinoma cell lines. It was determined that the mechanism of inhibition was apoptosis and the prevention of cell entry into the mitotic phase of the cell cycle.16, 40 Other reports have indicated that, in breast and prostate cancer cell lines, the degree of apoptotic cell death may be increased with the combination of mifepristone and tamoxifen.27, 38 It also is worth noting that progesterone therapy for endometrial cancer reportedly acts through the suppression of cell proliferation but not through apoptosis.28 Another aspect of growth inhibition induced by mifepristone may be related to vascular changes in the endometrium. It has been demonstrated in some cell types that progestins stimulate the expression of vascular endothelial growth factor and, thus, theoretically partially control cell growth or metastasis.41 The administration of mifepristone in vivo produced endometrial changes consistent with vascular stasis, vasoconstriction, vasoregession, and enhanced leukocytic infiltration.38

Two European clinical trials that used mifepristone (200 to 400 mg daily) in women with recurrent or progressive metastatic breast cancer resulted in a partial response or stable disease in 53% to 63% of women; unfortunately, however, the response duration was brief.31, 33 Romieu et al studied 22 patients with advanced breast cancer who failed tamoxifen and reported 12 partial responses or stabilization for 6 to 40 weeks; however, 82% of their patients progressed by 3 months.33 Response was correlated with positive PR status, as anticipated. Recently, a phase 2 trial using mifepristone 200 mg daily was completed in 44 patients with ovarian cancer. Thirty-four of those patients were evaluable, and there was a 26.5% objective response rate. Testing for positive ER/PR status was not performed in that trial. Again, the duration of response was short (range,1-4 months).32

Reasons for low response rates in this study may be related to the finding that the mechanism of action of mifepristone remains unclear. New studies have demonstrated that this SPRM may not work simply by modulating hormone receptors as reported previously. Several hypotheses now propose additional mifepristone activity on the expression of several proteins, including BAX, bcl-2, p53, and cyclooxygenase-2.15, 16 Supporting this, Kamradt et al demonstrated the restoration of p53 expression by mifepristone in dexamethasone-induced, radioresistant cervical carcinoma cells.42

In our population, mifepristone was well tolerated. The most frequently reported grade 1 and 2 tumor-related toxicities were anorexia, fatigue, and mood alterations (50%, 50%, and 58%, respectively). Eighty-three percent of patients had asymptomatic elevations of ACTH and cortisol. Mood alterations may have been related to changes in cortisol levels. No serious treatment-related adverse events occurred. Although the data suggest that patients' quality of life and symptoms improved over time, particularly those related to physical and functional well being and physical and psychologic symptoms, these observations were not statistically significant because of the small sample size.

Single-agent mifepristone was tolerated well and did not have a negative impact on quality of life. However, only a 25% stable disease response rate was recorded. Our findings suggest that single-agent mifepristone should not be used in this disease setting, because medroxyprogesterone acetate, megestrol acetate, leuprorelin, and combined megestrol acetate plus tamoxifen all would be expected to exhibit higher response rates (range, 9%-100%) in patients with grade 1, 2, and 3 tumors.9 However, a possible synergistic effect between mifepristone and chemotherapeutic agents was not addressed by this study. The results of this study underscore the need to continue developing novel treatments for patients with recurrent endometrial cancer. Further research into the exact mechanism of the cytotoxic effect of mifepristone on endometrial cancer lines is needed to explore further the benefit of mifepristone therapy in the treatment of endometrial cancer.

Conflict of Interest Disclosures

  1. Top of page
  2. Abstract
  6. Conflict of Interest Disclosures
  7. References

Supported in part by the University of Texas M. D. Anderson Cancer Center Gynecologic Specialized Program of Research Excellence for Uterine Cancers (CA098258).


  1. Top of page
  2. Abstract
  6. Conflict of Interest Disclosures
  7. References