Postmenopausal unopposed estrogens. Characteristics of use in relation to the risk of endometrial carcinoma.

Abstract

To determine whether treatment regimens for unopposed estrogens can be tailored so as to minimize the excess risk of endometrial cancer, results from 19 published studies of the association between unopposed estrogen use and endometrial cancer were compiled. We sought to examine the influence of duration of use, recency, dose, type of estrogen preparation, and periodic interruption of use on cancer incidence. Estrogen use for 5 years or longer was examined in 18 studies and was associated with a large increase in the risk of endometrial cancer in each one (range in relative risk, 1.8 to 36). Use for shorter durations also was observed to increase risk; however, among women who used estrogens for less than 6 months, any increased risk that may exist appears to be very small in size (six studies; range, 0.6 to 1.4). Risk consistently was seen to decrease with increasing time since cessation of use, although there is evidence from seven of eight studies that some residual excess risk remains long after estrogens have been discontinued. In each of 12 studies that examined the influence of dose, all dose levels of conjugated estrogens increased risk of endometrial cancer substantially. Four of five studies found no differences between oral synthetic estrogens and conjugated estrogens with respect to cancer risk, and all of eight studies found no difference between cyclic and continuous regimens. Based on our review, we conclude that apart from minimizing the duration of use, there is no way of taking unopposed postmenopausal estrogens that reduces their potential to cause endometrial cancer.