Long-term outcome of postmenopausal women with proliferative endometrium on endometrial sampling

The most common gynecologic malignancy is endometrial cancer. Endometrial hyperplasia (EH) is associated with increased risk of endometrial cancer in postmenopausal women. Early estrogenic effects produce proliferative changes in the endometrium of premenopausal women. Proliferative endometrium (PE), a common and physiologic finding in premenopausal women, is not expected to be present in the postmenopausal women. However, a number of studies have reported that up to 15% of the endometrial biopsies in postmenopausal women show PE. In contrast to EH, no association between PE with hyperplasia or cancer has been reported. The authors hypothesize that PE present in postmenopausal women may be associated with increased risk for hyperplasia or cancer. Data on long-term outcome of postmenopausal women with PE are scarce, and the clinical significance of PE in postmenopausal women is unclear. The aim of this retrospective cohort study was to determine the long-term outcome of postmenopausal women diagnosed with PE and to investigate the long-term risk of progression to EH or cancer in this group. Participants were women 55 years and older who underwent endometrial sampling as part of their routine clinical care at a large urban academic medical center between January 1997 and December 2008. The authors identified patients and obtained their medical information through use of Clinical Looking Glass software by applying the CPT code for endometrial biopsy/sampling. All available clinical outcome data were captured through February 2018. Women with a histological diagnosis of atrophic endometrium (AE) were used as controls. The proportions of women with PE and AE were compared with respect to the future development of EH or cancer. A subanalysis was performed for a subset of women presenting with postmenopausal bleeding. Univariable analysis was used to identify potential risk factors for EH or cancer. Multivariable logistic regression analysis was performed to identify independent risk factors. Of 962 women meeting inclusion criteria, 278 had PE and 684 had AE. Compared with women in the AE group, those in the PE group were younger (61.2 vs 64.5, P < 0.0001) and had a higher body mass index (BMI) (33.9 vs 30.6 kg/m², P < 0.0001). The length of surveillance was similar in both groups (11.9 vs 11.5 years, P = 0.27). Ethnicity was associated with PE (P = 0.01); there were more African American women in the PE group. Compared with the AE group, women in the PE group developed more EH or cancer (11.9% vs 2.9%, P < 0.0001), any endometrial cancer (5.8% vs 1.8%, P = 0.002), atypical EH (2.2% vs 0.4%, P = 0.02), and nonatypical EH (2.0% vs 0.7%, P = 0.001). The risk of endometrial cancer and EH remained similar in the PE group after excluding cases on hormonal therapy (12.2% vs 3%, P = 0.001). Logistic regression analysis showed that PE histology (odds ratio [OR], 3.89; 95% confidence interval [CI], 2.03-7.49; P < 0.0001), age older than 60 years (OR, 1.98; 95% CI, 1.03-3.82, P = 0.04), and BMI greater than 35 kg/m² (OR, 2.3; 95% CI, 1.09-4.83; P < 0.0001) remained significant and independent risk factors for progression to cancer. Sixteen percent of postmenopausal women who underwent endometrial sampling had PE and 11.9% of these developed endometrial cancer or hyperplasia during the follow-up period. There was a 4-fold greater incidence of developing endometrial cancer or hyperplasia among postmenopausal women with PE than those with AE. Because of the potential long-term risk of cancer progression, postmenopausal women with PE histology should have long-term monitoring. Prospective studies are needed to determine whether a treatment negating the risk of unopposed estrogen can prevent the development of EH or cancer. However, a large cohort with a similar long-term follow-up would be difficult to accomplish and is unlikely in the near future.