Hormone receptor expression patterns in the endometrium of asymptomatic morbidly obese women before and after bariatric surgery

https://doi.org/10.1016/j.ygyno.2013.12.005Get rights and content

Highlights

  • Occult endometrial hyperplasia is present in a notable minority of asymptomatic morbidly obese women.

  • Obese women with hyperplasia express higher levels of estrogen receptor, but these levels appear to decrease with weight loss.

Abstract

Objective

Obesity increases risk for endometrial neoplasia, but neither the pathophysiology nor the effects of weight loss on the risk are well established. We attempted to characterize the molecular profile of the endometrium of asymptomatic women with morbid obesity before and following bariatric surgery-induced weight loss.

Methods

59 asymptomatic, morbidly obese women underwent endometrial sampling before bariatric surgery; 46 (78%) of these returned one year later for re-biopsy (median weight loss of 41 kg). Duplicate samples from these specimens were scored for expression of estrogen receptor (ER), progesterone receptor (PR), androgen receptor (AR), and Ki-67 by two independent, blinded pathologists using an H-score [staining intensity (0–3) × (percent of tissue involved)].

Results

The prevalence of hyperplasia pre-operatively was 7% overall and 10% among patients not on an anti-estrogen. ER H-scores were similar before and after surgery overall (median 190 and 196 respectively, p = 0.82), but patients with hyperplasia had higher pre-operative H-scores (median 256, p < 0.001) and experienced greater H-score drops, than those without hyperplasia (− 112 vs + 50, p = 0.028). In two patients with persistent hyperplasia at one year, ER H-scores fell to levels that were similar to those without pathology. One patient who developed hyperplasia during the study period had a rising ER H-score. Patients with hyperplasia had higher median PR H-scores pre-operatively (284 vs 188, p = 0.01), which normalized through greater drops (75 vs 0, p = 0.053). AR H-scores dropped significantly after surgery (13 vs 2, p = 0.015), but were similar between patients with and without hyperplasia (p = 0.33). Weight loss did not affect Ki-67 proliferation index.

Conclusion

Asymptomatic morbidly obese patients have a high prevalence of occult hyperplasia, characterized by relatively high hormone receptor expression. These profiles appear to normalize with weight loss and in advance of pathologically identifiable changes. These data suggest a potential role for screening this population as well as the possibility that weight loss may be a valid treatment strategy for risk reduction.

Introduction

Increasing obesity in women is associated with a progressively worsening estrogen/progesterone imbalance [1]. The causal effect of unopposed estrogen on the induction of endometrial cancer is singularly well established through historical study of early hormone replacement efforts [2]; but there is increasing epidemiologic evidence indicating that obesity increases the risk of adenocarcinoma in other endocrine sensitive organs such as the breast, ovary, and prostate [3], [4], [5]. Simple estrogen imbalance, however, does not account for all of the increased risk attributable to obesity. Obese patients also have an increased risk for adenocarcinomas of the pancreas and colon which are not traditionally considered endocrine-sensitive, suggesting that alternate or multiple molecular pathways may be involved [6].

Despite a rising incidence and relative mortality of endometrial cancer, coinciding with a national epidemic of obesity, no prevention or screening strategy has proven cost-effective to date [7]. This is largely owing to a relatively low prevalence of endometrial pathology in the asymptomatic general population and a tendency for detection at early stages because of symptomatic bleeding. Recently however, our group has completed a natural history study of the effect of bariatric surgery-induced weight loss, in which we observed that the rate of occult hyperplasia in untreated, asymptomatic, morbidly obese women is up to 10% and that weight loss reduced but did not eliminate this risk [8]. This level of premalignant disease suggests that in morbidly obese women screening or proactive intervention may be cost effective.

Ultimately, effective screening efforts will require the treating clinicians to be able to stratify risk based on objective and reproducible measures, while prevention efforts will require an understanding of the natural history of disease progression and the range of potential treatment options. Multiple studies have demonstrated that the risk of coincident or subsequent endometrial cancer rises along a continuum of endometrial architectural complexity and cytologic atypia [9]. Despite these findings, traditional pathologic assessments have high inter-observer variability, and are increasingly unreliable at the lower end of the risk spectrum. Efforts to simplify the grading system using endometrial intraepithelial neoplasia (EIN) nomenclature have encouraging initial results, but have been slow to be adopted, and still rely heavily on microscopic, as opposed to molecular features [10].

The expression patterns of both the estrogen receptor (ER) and the progesterone receptor (PR) in normal endometria and to a lesser extent in malignant and premalignant endometria are well characterized [11]. However, the significance of the expression patterns in obese women, the implications of androgen receptor (AR) expression, and the effect of weight loss on hormone expression patterns remain unknown. The present study was undertaken to determine if morbidly obese patients with premalignant changes had detectably different expression profiles of markers known or suspected to be involved in the development of endometrial cancer; and to determine if weight loss affected these profiles.

Section snippets

Tissue acquisition

After obtaining institutional review board approval, 59 patients were enrolled in a prospective non-interventional trial to assess the effect of Roux-en-Y gastric bypass surgery on the prevalence of endometrial pathology. The details of this study are reported elsewhere; but briefly, all morbidly obese women who were asymptomatic by self report (no abnormal vaginal bleeding in the previous 6 months), had a negative pregnancy test or prior sterilization, and gave consent were eligible. We

Clinical results

59 morbidly obese women underwent initial sampling. Patient demographics are listed in Table 1, most patients were Caucasian, less than 50 years old, had good access to health care and relatively few medical co-morbidities. All patients underwent Roux-en-Y bypass and endometrial biopsy without overt complication, though in five pre-operative (8.5%) and one post-operative (2.2%) specimens there was insufficient tissue for diagnosis. Twelve months post-operatively the median (range) weight loss

Discussion

Our data indicate that obese women with hyperplasia have a different profile of hormone receptor expression from women without endometrial pathology. The differential expression pattern appears to resolve with weight loss, even in the absence of histological change; and in this sample, preceded the resolution of histological abnormalities in half of the four cases that ultimately resolved. The intricate balance of the hormonal influences, regulated through their respective receptors, is crucial

Conclusion

Morbidly obese women have a relatively high risk for harboring endometrial pathology even when asymptomatic. In some patients, both the histologic abnormalities and the high expression of ER and PR appear to normalize over one year in the absence of other medical interventions suggesting that the effects of bariatric surgery-induced weight loss may favorably influence a patient's risk of developing endometrial neoplasia. The high prevalence of occult disease in this population and the absence

Conflict of interest statement

The authors declare no conflicts of interest with regard to the results or interpretation of this data.

Acknowledgments

This research was supported by an American Cancer Society Mentored Research Award (#MRSG-10-079-01-CPPB) and by a grant from the Scaife Foundation, for which the authors are grateful.

References (20)

There are more references available in the full text version of this article.

Cited by (26)

  • Does weight loss in women with obesity induce regression of endometrial hyperplasia? A systematic review

    2023, European Journal of Obstetrics and Gynecology and Reproductive Biology
  • Prevention of endometrial cancer through lifestyle Interventions: A systematic review and synthesis

    2022, Gynecologic Oncology Reports
    Citation Excerpt :

    This can be visualised in Supplementary Table 2. The majority of included studies (7 of the 9 studies) specifically recruited women with obesity due to the strong association between endometrial cancer and obesity (Linkov et al. 2014; Linkov et al. 2012; Linkov et al. 2017; MacKintosh et al. 2019; Argenta et al. 2014; Yates et al. 2018). Lu et al recruited women between the ages of 25 and 50 with Lynch syndrome (defined by having a known mutation in MLH1, MSH2, or MSH6) a known high-risk population for endometrial cancer.

  • Obesity and reproduction: a committee opinion

    2021, Fertility and Sterility
    Citation Excerpt :

    Regardless, obesity is a well-known risk factor for endometrial hyperplasia and cancer (15). On a molecular level, the endometrium from women with obesity demonstrates increased steroid receptor staining and altered expression of other genes that is more pronounced in the presence of infertility (83, 84). Obesity has been associated with an increased risk of pregnancy loss (56, 58, 85–89) in most, but not all, studies (55, 57, 90).

  • Endometrial cancer and bariatric surgery: A scoping review

    2019, Surgery for Obesity and Related Diseases
    Citation Excerpt :

    In their case report, Bentio et al. [22] described a grade I endometrioid adenocarcinoma of the endometrium that failed treatment with a levonorgestrol intrauterine device and was found to regress at 18 months postsurgery. Estrogen and progesterone receptors in the endometrium also have been demonstrated to normalize postoperatively [23], while tumor suppressor gene expression has not demonstrated significant differences [24]. Metabolic profiles have also demonstrated favorable changes toward a decreased risk in endometrial cancer, demonstrating improved insulin response, decreased inflammation, and normalization of steroid hormones [20,25].

View all citing articles on Scopus
View full text