Randomized Double-Blind Trial of Estrogen Replacement Therapy vs Placebo in Stage I or II Endometrial Cancer

Abstract & Commentary

By Robert L. Coleman, MD, Associate Professor, University of Texas; M.D. Anderson Cancer Center, Houston. Dr. Coleman is on the speaker's bureau for GlaxoSmithKline, Bristol-Myers Squibb, and Ortho Biotech.

Synopsis: Although this incomplete study cannot conclusively refute or support the safety of exogenous estrogen with regard to risk of endometrial recurrence, it is noteworthy that the absolute recurrence rate (2.1%) and the incidence of new malignancy were low.

Source: Barakat RR, et al. Randomized double-blind trial of estrogen replacement therapy versus placebo in stage I or II endometrial cancer: a Gynecologic Oncology Group Study. J Clin Oncol. 2006;24:587-592.

The association between exogenous estrogen use and endometrial cancer development has been well documented. Nonetheless, the hazard of estrogen replacement therapy in women with a personal history of endometrial cancer has not been well studied and, despite anecdotal evidence of its safety, is largely discouraged among clinicians. In this regard, Barakat and colleagues from the Gynecologic Oncology Group (GOG) set out to quantify the risk/benefit of estrogen replacement in women with early stage endometrial cancer. Women undergoing primary extirpative surgery for stage IA-IIB (occult) endometrial cancer were randomized to either estrogen (premarin 0.625 mg) or to placebo. All women were to have an indication for estrogen therapy including vasomotor symptoms, vaginal atrophy, increased risk for cardiovascular disease or an increased risk for osteoporosis. The intent of therapy was 3 years administration. Strata were considered for stage based on differing survival factors. The primary end point was time to recurrence with all cause survival a secondary objective.

Based on anticipated rates of recurrence and loss to crossover, 2108 women were to be recruited. Over 66 months, 1236 women were randomized to the trial. Effects on accrual following the announcement of data from the Women's Health Initiative (WHI) caused the GOG to terminate the study early. Half of the cohort were randomized to each treatment group. Just 41% of patients on ERT and 50% on placebo remained compliant over the course of 3 years. Recurrence was lower than expected at 2.1%, which was countered by the GOG through closure to Stage IA patients. Nonetheless, just 9 of the 45 deaths observed on the trial were from endometrial cancer (0.7% of total population). There was no observed difference in recurrence or all-cause death between the treatment arms but the trial did not meet its accrual goals in order to infer this result in this population. New malignancy rates were low (1.3% for ERT, 1.6% for placebo) in both cohorts.


Endometrial cancer is the most common gynecologic malignancy diagnosed in the United States. Fortunately, most are confined to the uterus at presentation, which generally translates into a favorable survival profile compared to other malignancies of the genital tract. While the median age of diagnosis is older than 60, a sizeable fraction of patients are perimenopausal at diagnosis and suffer from significant estrogen deprivation following primary surgical management. Since estrogen use has been linked with the development of this disease, clinicians have been reticent to use ERT even in symptomatic women with a personal history of endometrial cancer based on the theoretical risk of "activating" dormant cells leading to recurrence. Retrospective studies have failed to confirm this suspicion.1-3 However, the issue of ERT use is relevant as the likelihood for long-term, cancer-free survival for these women is high and their exposure to other life-threatening factors related to estrogen deprivation is significant.

Data from the WHI caused a dramatic effect in the worldwide use of ERT. While the initial reports detailed outcomes based on the use of combined estrogen and progestin, the reduction in popularity heavily influenced the trial's accrual necessitating early termination. Subsequently, data from WHI with estrogen alone regimens have been published which confirmed increased risks of thromboembolic events but no increase in cardiovascular disease or breast cancer. Protection of osteoporosis was confirmed but no specific evaluation of the influence of this therapy on quality of life was detailed. Arguably, this is of primary importance in the population being studied. While the current trial cannot be used to support or refute the use of ERT in women with endometrial cancer, the data provided do suggest that recurrence in this setting is low; but the compliance with therapy is also low. Short-term use for symptomatic patients does not appear to be associated with a dramatic acceleration of recurrence although safety cannot be definitively inferred. Similar to the discussion of ERT use in women without cancer, it would be prudent for symptomatic women with an endometrial cancer history desiring ERT to carefully counsel them to the known risks/benefits of ERT use and limit usage to the shortest practical period.