Clinicopathological and molecular analysis of endometrial carcinoma associated with tamoxifen

Novel targets and their functions in the prognosis of uterine corpus endometrial cancer patients

Aberrant mRNA expression is implicated in uterine corpus endometrial carcinoma (UCEC) oncogenesis and progression. However, effective prognostic biomarkers for UCEC remain limited. We aimed to construct a reliable multi-gene risk model using gene expression profiles. Utilizing TCGA data (543 UCEC samples, 35 controls), we identified 1517 differentially acting genes. Weighted gene co-expression complex analysis (WGCCA), hub gene screening, and risk regression analysis (RRA) were employed to determine prognosis-related genes and construct the risk model. Nomograms visualized risk scores and receiver operator characteristic (ROC) curves assessed model performance. Seven novel prognosis-related hub genes (ANGPT1, ASB2, GAL, GDF7, ONECUT2, SV2B, TRPC6) were identified. The model's concordance index (C index) by multivariate Cox regression analysis was 0.79. ROC curves yielded AUCs of 0.811 (3-year) and 0.79 (5-year), demonstrating the model's efficacy in predicting UCEC survival. Our study proposes a promising seven-biomarker risk model for predicting UCEC prognosis, offering potential clinical utility.

Clinicopathological and molecular pathological characteristics in tamoxifen‑related endometrial cancer

Tamoxifen (TAM), a selective estrogen receptor modulator, is often used for long-term adjuvant endocrine therapy in patients with hormone receptor-positive breast cancer. TAM is known to increase the risk of endometrial cancer (EC); however, the mechanism has not yet been fully elucidated. Therefore, molecular genetic analysis of EC following TAM administration (TAM-related EC) was conducted. A total of 10 samples of TAM-related EC and 20 sporadic EC samples (as controls) were analyzed. Copy number variation analysis was conducted, microsatellite instability (MSI) status was assessed, and mismatch repair (MMR) protein expression was examined immunohistochemically. Copy number variation analysis revealed that KDR, NOTCH1, NTRK1, NTRK3 and PDGFRB were more frequently amplified in TAM-related EC (P=0.039, P<0.001, P=0.011, P=0.006 and P=0.035, respectively). In MSI analysis, 4 cases were classified as MSI-high (40%), which is a higher frequency compared with that among patients with sporadic EC (~10% in Japanese women). Loss of MMR proteins was confirmed in all MSI-high cases. In 1 MSI-high case, a benign lesion of hyperplasia prior to EC development was also MSI-high with loss of some MMR protein expression. Several genes were specifically amplified in TAM-related ECs. Furthermore, TAM-related ECs were frequently MSI-high. Further studies are required to be conclusive; however, the present findings may lead to a reduction of unnecessary gynaecological testing in clinical practice and also encourage the testing for MSI status for optimal individualized treatment.

The management of menopause in women with a history of endometriosis: a systematic review

BACKGROUND

Endometriosis is typically regarded as a premenopausal disease, resolving after natural or iatrogenic menopause due to declining oestrogen levels. Nonetheless, case reports over the years have highlighted the incidence of recurrent postmenopausal endometriosis. It is now clear that both recurrence and malignant transformation of endometriotic foci can occur in the postmenopausal period. Postmenopausal women are commonly treated with hormone replacement therapy (HRT) to treat climacteric symptoms and prevent bone loss; however, HRT may reactivate endometriosis and stimulate malignant transformation in women with a history of endometriosis. Given the uncertain risks of initiating HRT, it is difficult to determine the best menopausal management for this group of women.

OBJECTIVE AND RATIONAL

The aim of this study was to systematically review the existing literature on management of menopausal symptoms in women with a history of endometriosis. We also aimed to evaluate the published literature on the risks associated with HRT in these women, and details regarding optimal formulations and timing (i.e. initiation and duration) of HRT.

SEARCH METHODS

Four electronic databases (MEDLINE via OVID, Embase via OVID, PsycINFO via OVID and CINAHL via EbscoHost) were searched from database inception until June 2016, using a combination of relevant controlled vocabulary terms and free-text terms related to 'menopause' and 'endometriosis'. Inclusion criteria were: menopausal women with a history of endometriosis and menopausal treatment including HRT or other preparations. Case reports/series, observational studies and clinical trials were included. Narrative review articles, organizational guidelines and conference abstracts were excluded, as were studies that did not report on any form of menopausal management. Articles were assessed for risk of bias and quality using GRADE criteria.

OUTCOMES

We present a synthesis of the existing case reports of endometriosis recurrence or malignant transformation in women undergoing treatment for menopausal symptoms. We highlight common presenting symptoms, potential risk factors and outcomes amongst the studies. Sparse high-quality evidence was identified, with few observational studies and only two randomized controlled trials. Given this paucity of data, no definitive conclusions can be drawn concerning risk.

WIDER IMPLICATIONS

Due to the lack of high-quality studies, it remains unclear how to advise women with a history of endometriosis regarding the management of menopausal symptoms. The absolute risk of disease recurrence and malignant transformation cannot be quantified, and the impact of HRT use on these outcomes is not known. Multicentre randomized trials or large observational studies are urgently needed to inform clinicians and patients alike.

Molecular mechanisms of tamoxifen-associated endometrial cancer (Review)

Tamoxifen has been prescribed to millions of females for breast cancer prevention or treatment. However, tamoxifen is known to significantly enhance the risk of developing endometrial lesions, including hyperplasia, polyps, carcinomas, and sarcoma. Notably, tamoxifen-associated endometrial cancer often has a poor clinical outcome. Understanding the molecular mechanism of tamoxifen-induced endometrial cancer is essential for developing strategies that minimize tamoxifen’s effects on the endometrium without jeopardizing its breast cancer treatment effects. However, this understanding remains limited. Tamoxifen appears to mediate its effect on endometrial cells through estrogenic and non-genomic pathways, rather than introducing a genomic alteration as a carcinogen. Although tamoxifen functions as an agonist and promotes cell proliferation in endometrial cancer, it also displays antagonist activity towards some estrogen targets. Alterations in estrogen receptor-α and its isoforms, as well as the membrane associated estrogen receptor G protein-coupled receptor 30, have been observed with tamoxifen-exposed endometrial cells, and likely mediate the effects of tamoxifen on endometrial cancer cell proliferation and invasion. In addition, gene profile studies of short-term exposure to tamoxifen indicate that the majority of tamoxifen targets are tamoxifen-specific. However, the tamoxifen regulated gene targets that are involved in mediating the effects of long-term exposure to tamoxifen are not yet fully understood. Recent progress has indicated a potential role of unfolded protein response and mammalian target of rapamycin signaling in tamoxifen-associated endometrial cancer. In the future, studies focusing on long-term effects of tamoxifen exposure are required to understand the molecular mechanisms of tamoxifen-associated endometrial cancer.

Inducing malignant transformation of endometriosis in rats by long-term sustaining hyperestrogenemia and type II diabetes

This study aimed to induce malignant transformation of endometriosis in Sprague-Dawley rats by hyperestrogenemia and type II diabetes and evaluate its similarity with human disease in biological features. Rats with surgically induced endometriosis were randomized into two groups: those treated with estradiol (5 mg/kg three times/week after surgery), streptozotocin (25 mg/kg, 1 month after surgery), and high carbohydrate-and-fat feed (Es group); and those treated with placebo saline and standard feed (control group). All rats were randomly killed 2, 4, or 8 months after surgery. The endometriosis lesions and the corresponding eutopic endometria were subjected to morphological evaluation, TUNEL, and immunohistochemical analysis for the expressions of proliferating cell nuclear antigen, phosphatase and tensin homolog, phosphorylated protein kinase B, and phosphorylated mammalian target of rapamycin proteins. In the Es group, three cases (6.0%) of endometriosis showed atypical hyperplasia accompanied by simple hyperplastic eutopic endometria, and two cases (4.0%) of endometriosis showed endometrioid carcinoma accompanied by atypical hyperplastic eutopic endometria. In the Es group, the activity of organelles and the expressions of proliferating cell nuclear antigen, phosphorylated protein kinase B, and phosphorylated mammalian target of rapamycin increased, and the level of phosphatase and tensin homolog and TUNEL positivity decreased progressively in the order of endometriosis, atypical endometriosis, and malignant endometriosis. The same tendency was found in the corresponding eutopic endometria. The induced malignant endometriosis showed similarities with human disease in the pathological process and histomorphological and molecular biological features. The method is feasible. The malignant transformations of endometriosis and eutopic endometria may have correlations and similarities, but the former may suffer a higher risk of canceration.

MGMT promoter hypermethylation and K-RAS, PTEN and TP53 mutations in tamoxifen-exposed and non-exposed endometrial cancer cases

BACKGROUND

Tamoxifen has anti-oestrogenic and anti-tumour activity in the breast, but is oestrogenic and carcinogenic in the endometrium. It can induce experimental tumours by both hormonal and DNA-damaging mechanisms, but its carcinogenic mode of action in human endometrium remains unclear.

METHODS

We investigated whether an epigenetic mechanism, involving promoter hypermethylation of the gene for the DNA repair enzyme MGMT (O6-methylguanine DNA methyltransferase), was associated with K-RAS, TP53 and PTEN mutations in endometrial tumours from women treated with tamoxifen (TAM, n=30) or unexposed to the drug (EC, n=38).

RESULTS

There were significant (P<0.05) differences in tumour grade between the TAM and EC groups, with more favourable morphology in the latter. K-RAS mutations, predominantly G>A, occurred in small numbers in both groups. TP53 mutations were of mainly A>G, C>T and indel modifications in both groups, but more frequent in TAM cases. PTEN mutations dominated in EC tumours and were of the type that has large impact on protein function, such as indel or nonsense mutations. These observations alongside the mutational spectrum in PTEN suggest that the malignancies arise from different backgrounds, hence pointing to an effect of tamoxifen. Both groups displayed MGMT promoter hypermethylation. This coincided with mutations more frequently in the TAM (78%) than in the EC (50%) group, even though there were significantly (P<0.05) fewer mutations and methylations in TAM cases.

CONCLUSIONS

Although the difference in coincidence did not reach significance with the current sample size, the findings suggest that epigenetic processes may play a role in the way tamoxifen induces endometrial cancer.

Role of cyclins A and E in endometrial carcinogenesis in breast cancer patients under tamoxifen treatment

PURPOSE

The objective of our study was to determine the relevance of cyclins A and E overexpression in endometrial carcinogenesis in hormone receptor-positive breast cancer patients under tamoxifen therapy.

EXPERIMENTAL DESIGN

We assessed expression of cyclins A and E in Endometrial cytology samples collected from 36 ER and PR positive breast cancer patients; under tamoxifen treatment by using the Tao-brush non-invasive brushing cytology technique. Cyclins were detected in the collected samples by means of immuno-cytochemistry. The patients included in this study are a cohort of 36 breast cancer patients who were operated upon at the National Cancer Institute - Cairo University in the period from February 2006 to May 2008 and received tamoxifen (TAM) as part of their adjuvant treatment.

RESULTS

Cyclins A and E were expressed in 17 and 15 of the 36 collected endometrial cytology samples (47.2% and 41.6% respectively). Expression of cyclins A and E was highly correlated to Tamoxifen exposure duration (32 and 43 months respectively) p < 0.001. Tamoxifen median exposure duration was shortened to 21 months in cases showing positivity for either markers, while in cases showing positivity for both cyclins, the median exposure duration was longer (44.5 months) (p < 0.001). Neither cyclin A nor E was detected before median tamoxifen exposure duration of 11 months. Endometrial carcinoma cases had the longest Tamoxifen exposure duration (60 months).

CONCLUSION

Cyclins A and E expression is involved in the carcinogenesis of endometrium in women with breast cancer and under tamoxifen-treatment. Follow up of the patients using these 2 markers is highly recommended starting from the 12th month.

Prophylactic use of levonorgestrel-releasing intrauterine system in women with breast cancer treated with tamoxifen: a randomized controlled trial

OBJECTIVE

To estimate the rate of endometrial pathology with the prophylactic use of levonorgestrel-releasing intrauterine system in women with breast cancer treated with tamoxifen.

METHODS

This was a randomized contro-lled trial of 129 Chinese women who attended a university hospital in Hong Kong and required adjuvant tamoxifen for breast cancer after the completion of postoperative radiotherapy and chemotherapy. Women were randomized to treatment (prophylactic levonorgestrel-releasing intrauterine system insertion before the commencement of tamoxifen) or control group. The uterine cavity was examined by hysteroscopy and endometrial sampling before the commencement of tamoxifen and at 12, 24, 45, and 60 months afterward. Any endometrial polyps or submucosal fibroids were resected through hysteroscopy at each assessment and specimens were sent for histologic confirmation.

RESULTS

A total of 94 women completed 5-year follow-up. There was no significant difference in the occurrence of submucosal fibroids (1 [1.8%] compared with 2 [3.4%]) and endometrial hyperplasia (both 0) in the treatment and control groups, respectively. Levonorgestrel-releasing intrauterine system significantly reduced de novo endometrial polyps (hazard ratio 0.19, 95% confidence interval 0.07-0.48) over the course of 5 years on an intention-to-treat basis. There was no statistically significant increase in breast cancer recurrence rate (10 [17.2%] compared with 6 [10.0%]) or cancer-related deaths (6 [10.3%] compared with 5 [8.3%]) in the treatment group, but the study was underpowered in this regard.

CONCLUSIONS

Prophylactic levonorgestrel-releasing intrauterine system prevents de novo endometrial polyps in women using tamoxifen. However, its role in the prevention of endometrial hyperplasia and adenocarcinoma as well as its effect on risk of breast cancer recurrence remain uncertain.

CLINICAL TRIAL REGISTRATION

Chinese Clinical Trial Registry, http://www.chictr.org/en/, ChiCTR-TRC-09000625.

LEVEL OF EVIDENCE

I.

New and potential clinical applications of KRAS as a cancer biomarker

IMPORTANCE OF THE FIELD

KRAS mutation is the most common oncogenic alteration in various human cancers. Recently, KRAS has emerged as an important predictive biomarker in common malignancies such as metastatic colorectal cancer (mCRC) and non-small cell lung cancer (NSCLC). This work aims to discuss the clinical impact of the KRAS mutation status on state-of-the-art treatment approaches, including epidermal growth factor receptor (EGFR)-targeted therapies.

AREAS COVERED IN THIS REVIEW

This review considers the potential of KRAS to serve as a diagnostic, prognostic or predictive biomarker in various cancers, including those of the lung, colon/rectum, pancreas, ovary and endometrium.

WHAT THE READER WILL GAIN

KRAS mutations in mCRC and NSCLC primary tumors predict resistance to EGFR-targeted therapy. In pancreatic cancer, KRAS may prove useful as a diagnostic biomarker to screen for early neoplasia. Furthermore, quantitative KRAS mutation analysis could have the potential to distinguish pancreatic cancer from other conditions such as chronic pancreatitis. With respect to ovarian and endometrial cancer, further studies should focus on determining reliable biomarkers for predicting response to EGFR-targeted therapy. Besides EGFR inhibition, KRAS may also serve as a diagnostic and predictive biomarker for evolving therapies directed against mutant RAS proteins.

TAKE HOME MESSAGE

KRAS has been recognized as an outstanding predictive biomarker to select mCRC and NSCLC patients for EGFR-targeted therapies; however, multi-determinant approaches including other molecular markers should facilitate the identification of patients likely to respond to such therapies.

Mutual contribution of Pten and estrogen to endometrial carcinogenesis in a PtenloxP/loxP mouse model

OBJECTIVE

An association between estrogen and the risk of type 1 endometrial carcinoma, which shows frequent mutations in the Pten tumor suppressor gene, has consistently been found in many studies. However, such tumors usually arise in perimenopausal or postmenopausal women with decreased serum estrogen levels. This study aimed to reveal the contributions of estrogen to endometrial carcinogenesis in a mouse model of endometrial carcinoma initiated by conditional targeting of Pten.

METHODS

The Cre-loxP system was used to achieve Pten inactivation within mouse endometrial epithelium. We delivered a recombinant adenovirus vector expressing Cre recombinase to the endometrial cavity of the Pten mice that had been ovariectomized at 10 weeks old. Mice were subcutaneously injected with 17β-estradiol (E2) or vehicle, followed by injection of the adenovirus. Two weeks after adenovirus injection, the entire endometrium was analyzed.

RESULTS

Mice that did not receive E2 injection notably developed endometrial neoplasia, complex atypical hyperplasia, or carcinoma (7/8, 87.5%). In contrast, hyperplastic but nonneoplastic endometrium was observed in E2-treated mice. In these E2-treated mice, immunohistochemistry revealed that Pten-null glandular epithelial cells clonally proliferate among the hyperplastic endometrium.

CONCLUSIONS

The results of this study suggest that estrogen clonally proliferates Pten-null epithelial cells together with surrounding cells, and depletion of estrogen induces predominant growth of Pten-null cells with estrogen-independent capabilities, resulting in abnormal structure of the glandular cells and subsequent neoplasia. This phenomenon might explain why the incidence of human endometrial carcinoma increases with perimenopausal or postmenopausal status, which represents declining ovarian function. Our model mice have partially resolved the issue of endometrial Pten- and estrogen-related carcinogenesis and have potential to represent a valuable tool for developing novel therapies against this carcinoma.