Event In Endometrial Carcinogenesis Research Articles

Explore the alterations of downstream molecular pathways caused by ARID1A mutation/knockout in human endometrial cancer cells.

As one of the most common gynecologic malignancies, endometrial cancer (EC) is driven by multiple genetic alterations that may be targeted for treatments. AT-rich interaction domain 1A (ARID1A) gene mutations were reported as early events in endometrial carcinogenesis. To explore the alterations of downstream molecular pathways caused by ARID1A mutations and the associated therapeutic implications, we edited ARID1A gene in human endometrial cancer cell line Ishikawa using the Clustered Regularly Interspaced Short Palindromic Repeats and CRISPR-Associated Proteins (CRISPR/Cas9) technology. We successfully constructed a stable Ishikawa cell line with a confirmed 10bp deletion on the ARID1A gene, which resulted in a code-shift mutation and gene knockout. Compared with unedited wild-type cells, ARID1A knockout (KO) led to reduced apoptosis, accelerated transformation from G0/G1 to S phase, and enhanced cell proliferation. ARID1A deficiency would reduce the protein levels of p21, caspase 7, and caspase 9 in Ishikawa endometrial cancer cells compared with the wild-type cells. In addition, ARID1A KO resulted in high levels of microsatellite instability (MSI-H). Moreover, transcriptomic analyses showed that ARID1A KO can lead to activated phosphoinositide 3-kinase (PI3K)/protein kinase B (Akt) signaling. Furthermore, experimental analyses demonstrated that ARID1A KO cells had reduced expression of genetic instability-associated markers mutL homologue 1 (MLH1) and progesterone receptor B (PR) and increased p-Akt expression. These findings support further exploration of ARID1A as a therapeutic target for EC and provide insight into developing more effective treatments in EC, such as the combinatory use of immune checkpoint inhibitors.

Alterations of UHRF Family Expression and UHRF1/ICBP90 Inhibits Phosphatase and Tensin Homolog Expression in Endometrial Cancer

Introduction: The altered protein expression of inverted CCAAT box-binding protein of 90 kDa/ubiquitin-like with PHD and RING finger domains 1 (ICBP90/UHRF1) and Np95-like ring finger protein (NIRF)/UHRF2, which belong to the ubiquitin-like with PHD and RING finger domains (UHRF) family, is linked to tumor malignancy and the progression of various cancers. To determine the role of NIRF and ICBP90 in endometrial tumorigenesis, we evaluated ICBP90 and NIRF expression levels in endometrial cancers. Also molecular alterations of phosphatase and tensin homolog (PTEN) expression are the important event for endometrial carcinogenesis; therefore, we investigated the involvement between ICBP90 and PTEN expression. Methods: We used Western blot for NIRF, ICBP90, and PTEN expression, mutation analysis of NIRF gene, and immunohistochemical staining for the expression of NIRF and ICBP90. For immunohistochemical staining, we examined atypical endometrial hyperplasia, endometrial cancers, and noncancerous samples. Results: Our data showed that the reduced expression of NIRF and overexpression of ICBP90 occurred in atypical endometrial hyperplasia and endometrial cancer compared to the normal endometrium. The decrease in NIRF expression was significantly correlated with histological grade. Expression of ICBP90 was high, especially in the peripheral margin of a cancer nest. Western blot analysis of endometrial cancer cell lines referred an opposite correlation between ICBP90 and PTEN expression. Conclusion: Our findings suggested that continually overexpressed ICBP90 may contribute to the inhibition of PTEN expression, which is a frequent and important event in endometrial carcinogenesis. We propose that the reduced NIRF expression and ICBP90 overexpression is an early event in endometrial carcinogenesis; thus ICBP90 may be useful as a therapeutic target in this disease.

IMMUNOHISTOCHEMICAL STUDY ON PTEN AND CYCLIN D1 IN NON-NEOPLASTIC AND NEOPLASTIC ENDOMETRIAL LESIONS

Abstract
 Introduction: Endometrial carcinoma (EC) is the most common gynaecological malignancy in developed countries and has been classified into two groups, type 1 and type 2. Type 1 or endometrioid endometrial carcinomas (EECA) accounts for 80% of EC and are thought to develop following a continuum of premalignant lesions ranging from endometrial hyperplasia without atypia (EH) and atypical hyperplasia (AH). PTEN (phosphatase and tensin homolog), a tumor suppressor gene is commonly inactivated in 83 % of endometrioid carcinoma and 55% of precancerous lesions. Cyclin D1, a cell cycle regulator is overexpressed in about 40% of endometrial carcinomas.
 Aim: To study the expression of PTEN (Phosphatase and tensin homolog) and Cyclin D1 in non-neoplastic and neoplastic endometrial lesions by immunohistochemistry (IHC).
 Methods: A 2 year cross-sectional study (September 2017 to August 2019) on 115 endometrial samples was done in the Department of Pathology, RIMS. Histomorphological features and IHC expression of PTEN and Cyclin D1 in the various endometrial lesions were studied and evaluated, data collected in IBM SPSS Statistics 21 was statistically analyzed using Chi - square and Fisher’s Exact test.
 Results: Out of the 115 cases, 47(40.9%) were diagnosed as benign proliferative endometrium, 20(17.4%) benign secretory endometrium, 21(18.3%) hyperplasia without atypia, 15(13.0%) atypical hyperplasia and 12(10.4%) endometrial carcinoma with an age group spanning from 26-68 years (mean age = 46.4). Following IHC staining, 91.7%(11/12) and 83.3%(10/12) cases of EC and 80%(12/15) and 73.3%(11/15) cases of AH showed complete loss of PTEN expression and Cyclin D1 overexpression, respectively when compared to other benign lesions and was statistically significant (p < .001).
 Conclusion: Loss of PTEN and Cyclin D1 overexpression was seen in a significant number of EECA and AH, suggesting both as an early event in endometrial carcinogenesis. Therefore, we propose the use of PTEN and Cyclin D1 immunostaining as an adjunct to histopathological diagnosis as it may be informative in the identification and further management of premalignant endometrial lesions that are likely to progress to carcinoma
 Keywords: PTEN, Cyclin D1, endometrial hyperplasia, endometrial carcinoma, endometrioid endometrial carcinomas.

Expression of Cyclin D1 in Hyperplasia and Carcinoma of Endometrium and Its Correlation with Histologic Grade, Tumor Type, and Clinicopathological Features.

Background Endometrial carcinoma is often preceded by characteristic histopathologic lesions known as endometrial hyperplasia. Estrogen, p53, PTEN, and overexpression of cyclin D1 appear to be involved in the development of endometrial carcinogenesis. Design We evaluated and compared the expression profile of cyclin D1 expressions in 50 endometrial samples submitted as either endometrial curetting ( n = 34) or hysterectomy ( n = 16) specimens, which were diagnosed as simple hyperplasia ( n = 10), complex hyperplasia ( n = 06), atypical hyperplasia ( n = 04), and endometrial carcinoma ( n = 20). Ten cases of normal proliferative and secretory endometrium were selected as controls. Breast cancer with known cyclin D1 expression was selected as a positive control in each immunohistochemistry run. Results Cyclin D1 was significantly overexpressed in glands with complex hyperplasia and endometrial adenocarcinoma compared with proliferative or secretory endometrium and simple hyperplasia. A statistical difference was found in the extent of cyclin D1 positivity of simple hyperplasia and carcinoma of the endometrium ( p < 0.005). No statistical difference was seen between complex hyperplasia and carcinoma and clinicopathologic parameters in endometrioid carcinomas. All cases of clear cell carcinoma and serous carcinoma showed cyclin D1 immunoreactivity. Significant statistical difference was seen between cyclin D1 expression and only one clinicopathologic parameter, i.e., menopausal status in endometrial carcinomas Conclusion Cyclin D1 over expression may be an early event in endometrial carcinogenesis and cyclin D1 over expression may be an informative biomarker to recognize subsets of endometrial lesions that may be precancerous and therefore amenable to surgical therapy.

Comparative characteristics of the transcriptional activity of CDH1, CTNNB1, VEGFA genes and expression of proteins E-cadherin, β-catenin and VEGFA, coded by these genes in metastatic and non-metastatic endometrioid endometrial carcinoma

The violation of Wnt-signaling pathway and cyclic neoangiogenesis is early events in endometrial carcinogenesis. In this process an important role is played by epigenetic modifications and mutations of CDH1, CTNNB1, VEGFA genes, followed by expression violations of E-cadherin, β-catenin, VEGFA encoded by them. The aim . To study the expression levels of mRNA of CDH1, CTNNB1, VEGFA genes and the expression levels of E-cadherin, β-catenin, VEGFA in non-metastatic endometrioid endometrial carcinoma (EEC) and in EEC with metastases to regional lymph nodes. Materials and methods. Immunohistochemical study of invasive рТ 1-4 N 0 M x endometrioid endometrial carcinoma (EEC) (n=56; age – 42–83 years), invasive рТ 1-4 N 1-2 M x EEC (n=30; age – 48–79 years), samples of normal proliferative endometrium (PE) (n=30; age – 41–62 years) was performed. Molecular-genetic study of invasive рТ 1-4 N 0 M x EEC (n=10; age – 45–67 years), invasive рТ 1-4 N 1-2 M x EEC (n=10; age – 44–63 years), samples of PE (n=10; age – 46–59 years) was performed. Results. EEC is characterized by the lower mRNA expression level of CDH1 gene and by the higher mRNA expression levels of CTNNB1, VEGFA genes in comparison with the PE. E-cadherin expression level is 32.6 % lower in non-metastatic EEC and also is 58.10 % lower in metastatic EEC in comparison with the PE. β-catenin expression level is 32.56 % lower in non-metastatic EEC and also is 58.11 % lower in metastatic EEC in comparison with the PE. VEGFA expression level is 27.72 % higher in non-metastatic EEC and also is 17.87 % higher in metastatic EEC in comparison with the PE. There is lower β-catenin expression level in metastatic EEC in comparison with non-metastatic EEC. Non-metastatic and metastatic EEC is characterized by the direct medium connections between the mRNA expression levels of CTNNB1 and VEGFA genes (γ=0.44), between the mRNA expression level of CTNNB1 gene and VEGFA expression level in the tumor (γ=0.37). Conclusions . These changes contribute to tumor growth, invasion and metastasis.

Abstract 189: Protein kinase C alpha suppresses AKT activation in endometrial cancer

Abstract Endometrial cancer (EC) is the most common gynecological malignancy and the fourth leading cancer in women in the United States. Hyperactivation of the PI3K/AKT pathway is observed in ∼90% of EC cases and this pathway is the most prominent driver of endometrial carcinogenesis. Mutations in multiple PI3K/AKT pathway components (including PTEN, PIK3CA, PIK3R1, AKT) often coexist in EC, indicating that EC development requires full, unopposed activation of PI3K/AKT signaling. In this study, we identified PKCα as a negative regulator of AKT activation that suppresses endometrial carcinogenesis. Our analysis revealed that PKCα protein and mRNA are reduced or lost in 50% of EC cell lines and ∼60% of human ECs. PKCα deficiency was also observed in hyperplastic endometrial lesions in murine models with allelic knock-in of mutant Pten (ptenΔ4-5/+, ptenC124R/+ and ptenC129E/+) or endometrial-specific Pten deletion (ptenpr-/-, ptenltf-/-), indicating that loss of PKCα is an early event in endometrial carcinogenesis. Mechanistic analysis has further revealed that loss of PKCα is independent of changes in PI3K/AKT signaling; however, the enzyme acts as a potent negative regulator of the PI3K/AKT pathway in EC cells. PKC agonists reduced AKT phosphorylation/activity in EC cells that retain PKCα, via a PP2A-dependent mechanism. Conversely, PKCα knockdown led to increased AKT phosphorylation. Notably, while PKC agonists did not affect AKT in EC cells with loss of PKCα, exogenous expression of the enzyme in these cells restored the inhibitory effect of the agonists on PI3K/AKT signaling. Collectively these data indicate that PKCα has a restraining effect on PI3K/AKT signaling in EC cells, providing a basis for its loss during endometrial tumor progression. Physiological relevance of these findings was established using cell lines, patient samples and mouse models. PKCα restoration in human EC cells decreased expression of oncogenes such as cyclin D1 and Id1 and abolished their anchorage-independent growth, supporting a tumor suppressive role for the isozyme in this tissue. Analysis of PKCα expression in patient specimens determined that loss of PKCα expression correlates with high grade disease. Finally, in mice heterozygous for mutant Pten, PKCα knockout led to a 3-fold increase in endometrial tumor burden at three months, confirming a role of PKCα as a tumor suppressor in the endometrium. Taken together, our results provide evidence that PKCα 1) negatively regulates AKT signaling and the expression of important oncogenes in EC cells; and 2) functions as tumor suppressor in the endometrium whose loss is associated with disease progression. Thus, PKCα signaling represents a promising biomarker for risk stratification in early stage disease and may provide insight into therapeutic strategies for late stage disease. Supported by NIH grants CA036727, CA016056, and DK60632. Citation Format: Alice H. Hsu, Kathryn J. Curry, Kang-Sup Shim, Peter Frederick, Carl D. Morrison, Baojiang Chen, Subodh M. Lele, Takiko Daikoku, Sudhansu K. Dey, Gustavo Leone, Adrian R. Black, Jennifer D. Black. Protein kinase C alpha suppresses AKT activation in endometrial cancer. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 189.

Immunoexpression of Cyclin D1 and PTEN in Various Endometrial Pathologies.

To determine the expression of cyclin D1 and PTEN (phosphatase and tensin homolog) in endometrial hyperplasias and neoplasias. Analytical study. The study was conducted at BMSI, JPMC, Karachi, from January 2008 to December 2012. Analysis of endometrial samples, comprising of hysterectomies and curettage, was carried out. Immunohistochemical staining was done for PTEN and cyclin D1 expression. Fifty-three endometrial samples including 23 endometrial carcinomas, 6 complex hyperplasias with atypia, 14 complex hyperplasias without atypia, 6 simple hyperplasias without atypia and 4 normal proliferative endometrium were analyzed. Fifty-two percent (12 out of 23) and 48% (11 out of 23) cases of endometrial carcinomas showed complete loss of PTEN expression and cyclin D1 over expression, respectively. Five (5 out of 6) cases of complex hyperplasias with atypia and 64.28% (9 out of 14) cases of complex hyperplasia without atypia showed complete loss of or diminished expression of PTEN whereas 66.66% (4 out of 6) cases of endometrial hyperplasia with atypia and 50% (7 out of 14) cases of endometrial hyperplasia without atypia showed cyclin D1 overexpression (p &amp;lt; 0.001). Loss of PTEN, expression and cyclin D1 overexpression was seen in a significant number of well differentiated endometrial adenocarcinomas and complex hyperplasias with atypia, suggesting both as an early event in endometrial carcinogenesis.

PTEN expression in benign human endometrial tissue and cancer in relation to endometrial cancer risk factors.

Clonal loss of PTEN expression occurs frequently in endometrial carcinoma and endometrial hyperplasia. Limited data from immunohistochemical studies suggest that PTEN-null appearing endometrial glands are detectable in women without pathologic abnormalities, but the relationship of PTEN expression to endometrial cancer risk factors has not been extensively explored. We evaluated relationships between endometrial cancer risk factors and loss of PTEN expression in a set of benign endometrial samples prospectively collected from women undergoing hysterectomy and in endometrial cancer tissues from a population-based case-control study. We used a validated PTEN immunohistochemical assay to assess expression in epidemiological studies designed to assess benign endometrium [Benign Reproductive Tissue Evaluation Study (n = 73); Einstein Endometrium Study (n = 19)], and endometrial cancer [Polish Endometrial Cancer Study (n = 148)] tissues. Associations between endometrial cancer risk factors (collected via study-specific risk factor questionnaires) and PTEN expression in endometrial tissues were determined using Fisher's exact tests. PTEN loss was detected in 19% of benign endometrial tissues versus 55% in endometrial cancers. NSAID use was statistically significantly associated with PTEN loss in the benign endometrium (p = 0.02). Our data demonstrate that PTEN loss is detectable in endometrial tissues that are benign and malignant, with substantially more frequent loss in endometrial cancer compared with benign endometrium. However, alterations in expression were unrelated to most risk factors in this analysis, except for the association with NSAID use, which may represent a chance finding or reverse causality among patients with endometriosis who may have PTEN pathway abnormalities in eutopic endometrium. Further evaluation of factors associated with PTEN loss and long-term follow-up of women with PTEN-null endometrial glands may be useful in understanding early events in endometrial carcinogenesis.

Aberrant activation of hedgehog signaling pathway contributes to endometrial carcinogenesis through β-catenin

The hedgehog and Wnt signaling pathways play important roles in human cancers with possible interaction. This study aimed at analysis and correlation of the expression of Gli1, a transcriptional factor and target gene of hedgehog signaling pathway, with clinicopathological parameters and expression of β-catenin, an important member of the Wnt pathway, in normal, hyperplastic and malignant endometrium. Immunohistochemical study on 15 normal endometrium, 14 simple and complex hyperplasia without atypia, 37 atypical complex hyperplasia and 80 endometrial cancers showed significant Gli1 overexpression and β-catenin nuclear immunoreactivity in endometrial cancers and atypical endometrial hyperplasia when compared with normal endometrium (P<0.05). Overexpression of Gli1 in endometrial cancers correlated with well-differentiated histological grade (P<0.001), non-myometrial invasion (P=0.004) and superficial myometrial invasion (P=0.041). β-Catenin nuclear immunoreactivity was also associated with well-differentiated histology (P=0.013). Gli1 overexpression positively correlated with β-catenin nuclear immunoreactivity in atypical complex hyperplasia (P=0.013) and endometrial carcinoma (P=0.017). Similar Gli1 and β-catenin protein expression pattern was observed in normal and endometrial cancer cell lines by western blotting. We further showed a complex formation between Gli1 and β-catenin protein in endometrial cancer cell lines in an immunoprecipitation study. Ectopic overexpression of Gli1 into endometrial cancer cells led to reduced expression of β-catenin in cell cytoplasm and increased expression of β-catenin in the nuclei. In summary, overexpression of Gli1 was an early event in endometrial carcinogenesis. Aberrant activation of hedgehog pathway may play important roles in endometrial cancer through β-catenin nuclear accumulation.

Mutations of the KRAS oncogene in endometrial hyperplasia and carcinoma.

The aim of this study was to examine the prevalence and clinicopathological significance of KRAS point mutation in endometrial hyperplasia and carcinoma. We analysed KRAS in 11 cases of complex atypical hyperplasia and in 49 endometrial carcinomas using polymerase chain reaction associated with restriction fragment length polymorphism (PCR-RFPL). Point mutations at codon 12 of KRAS oncogene were identified in 7 of 49 (14,3%) tumor specimens and in 2 of 11 (18,2%) hyperplasias. No correlation was found between KRAS gene mutation and age at onset, histology, grade of differentiation and clinical stage. We conclude that KRAS mutation is a relatively common event in endometrial carcinogenesis, but with no prognostic value.

High frequency of RASSF1A and RARb2 gene promoter methylation in morphologically normal endometrium adjacent to endometrioid adenocarcinoma

To identify a DNA methylation signature of endometrioid carcinoma of the endometrium (EEC) in the early stages of endometrial carcinogenesis. Archival biopsy specimens of 39 EECs, 14 cases of atypical hyperplasia (AH), 11 histologically normal endometrial tissues adjacent to EECs and 24 normal control endometrial samples were retrieved. The cases were tested by quantitative methylation-specific polymerase chain reaction with primers hybridizing in the promoter regions of five genes frequently methylated in human cancer (RASSF1A, RARb2, P16, MGMT and GSTPi). Twenty-nine of 39 (74%) EECs and 7/14 (50%) AHs were methylated for the RASSF1A gene, whereas 17/39 (44%) EECs and 6/14 (43%) AHs were positive for the methylation of the RARb2 gene. No significant results were obtained for the other genes (P16, MGMT and GSTPi). Interestingly, 4/11 (36%) and 6/11 (55%) histologically normal endometrial tissues adjacent to EEC showed, respectively, RASSF1A and RARb2 gene methylation. Furthermore, these 11 specimens were microsatellite stable and showed similar proliferative, cell cycle and apoptotic mean labelling indices as the normal endometrial control tissues. Promoter region methylation of RASSF1A and RARb2 genes is an early event in endometrial carcinogenesis.

RASSF1A methylation and K-ras and B-raf mutations and recurrent endometrial cancer

Aberrations in mediators of Ras signaling may increase the risk of developing recurrent endometrial carcinoma. Primary tumors of patients with (n = 44) and without (n = 44) recurrent stage I endometrioid endometrial carcinoma were compared regarding the presence of K-ras mutations (codons 12 and 13), B-raf mutations (V599), and RASSF1A gene promoter methylation. K-ras mutations were present in 18% of the patients independent of recurrent disease. No B-raf mutations were found. RASSF1A methylation was demonstrated in 85% of endometrial carcinomas, independent of recurrence. The presence of K-ras mutations and RASSF1A promoter methylation were not related, either directly or inversely. Analysis in premenopausal endometrial carcinomas demonstrated K-ras mutations in 40%, no B-raf mutations, and RASSF1A promoter methylation in 70% of the cases. RASSF1A methylation was also observed in samples of cyclic (n = 14), hyperplastic (n = 8), and atrophic (n = 13) endometrial tissues in 21%, 50% and 38%, respectively. RASSF1A methylation was observed in a high frequency in endometrioid endometrial carcinoma whereas K-ras and B-raf mutations were observed in a low frequency. No association was observed with the development of recurrent disease. High-frequency RASSF1A methylation in premenopausal carcinomas and an increased frequency in endometrial hyperplasia indicate that this may be an early event in endometrial carcinogenesis.

Expression of Tumor-Associated Differentially Expressed Gene-14 (TADG-14/KLK8) and Its Protein hK8 in Uterine Endometria and Endometrial Carcinomas

To clarify the biological behavior of TADG-14/KLK8, we investigated TADG-14/KLK8 mRNA by semiquantitative RT-PCR and hK8 expression by immunohistochemistry using 37 normal endometria and 44 endometrial carcinoma tissues. TADG-14/KLK8 mRNA expression levels were significantly higher in proliferative compared to secretory phase endometria (p = 0.0143). Levels of TADG-14/KLK8 mRNA expression correlated with hK8 protein levels. hK8 was detected in 73.3% (11/15) of endometria with a significantly higher detection rate in the proliferative compared to secretory and atrophic phase endometria (p = 0.0002). High expression of hK8 was found in 61.4% of endometrial carcinomas compared to 35.1% of endometrial tissue samples (p = 0.0187). hK8 expression was significantly higher in stage I (p = 0.0433, 0.0038) and grade 1/2 (G1/2) of the tumors (p = 0.0195, 0.0044). We suggest that expression of TADG-14/KLK8may be regulated by sex steroid hormones in endometria. Our results indicate that elevated TADG-14/KLK8 expression is an early event in endometrial carcinogenesis, and may potentially serve as a useful early biomarker for the detection of endometrial carcinomas in menopausal women.

Consistent absence of BRAF mutations in cervical and endometrial cancer despite KRAS mutation status

Background. Mutational activation of KRAS and BRAF proto-oncogenes contributes to the development of many human cancers. Current research on gynecological cancer and specifically in cervical and endometrial cancer is focused on the mechanisms of their mutational activation. Objectives. In view of the paucity of data on their mutation frequency and the status of BRAF in these two types of gynecological cancer, we performed a systematic molecular study in 114 clinically and histologically well-defined malignant tumors of uterine cervix and endometrium and correlated the mutation status of KRAS and BRAF with the age at diagnosis and with tumor grade, stage or histological type. Methods. Direct sequence analysis of the PCR products of KRAS and BRAF genes was used to screen for known activating mutations. Results. In 67 cases of endometrial cancer, six KRAS mutations (8.9%) were found, four at codon 12 (5.9%) and two at codon 13 (2.9%), while no mutation was detected at codon 61. Most of the mutations occurred in surgical stage I and in the endometrioid adenocarcinoma subtype. We also detected three KRAS point mutations (6.3%) in the 47 cervical cancer samples, two at codon 12 (4.2%) and one at codon 13 (2.1%), while there was no mutation at codon 61. On the contrary, no mutation was identified in BRAF exon 15 for either endometrial or cervical cancer samples at position V600, which represents the most frequently mutated site of BRAF in human cancer. There was no association between KRAS mutations with either histological type, tumor grade or stage. Interestingly, however, KRAS mutation status in endometrial cancer was strongly associated with increased age at diagnosis ( P < 0.001). Conclusions. Our data document (a) the absence of BRAF mutations in cervical and endometrial cancer, despite the mutation status of KRAS, (b) suggest that KRAS mutations reflect an early event in endometrial carcinogenesis and (c) imply that BRAF activation is involving alternative pathways in these two types of cancer.

Cancer history and loss of MSH2 and MLH1 protein expression in patients with endometrial hyperplasia.

In order to evaluate the hereditary background of endometrial hyperplasia patients in relation to protein expression of DNA mismatch repair genes, we evaluated 69 patients with endometrial hyperplasia and 18 patients with normal endometrium having both a personal and family history of cancer (two hereditary nonpolypoid colorectal cancer (HNPCC) patients). We obtained personal and family histories of cancer for all patients. MSH2 and MLH1 protein expression was investigated by immunohistochemical methods. In the endometrial hyperplasia patients, 11 had personal histories and 40 had family histories of cancer. Among the 11 endometrial hyperplasia patients with a personal history of cancer, most cancers were breast or colorectal cancers (82%). In the 40 patients with a family history of cancer, colorectal cancer (33%) was the most frequent. The incidence of loss of expression of MSH2 and/or MLH1 protein in endometrial hyperplasia patients with personal (64%) or family (40%) histories was significantly higher than that in patients without such history (no personal: 21% and no family: 10%; P = 0.0035 and 0.0065). No protein loss was detected in any of the cases with normal endometrium having either a personal or family history of cancer. Our results suggest that a portion of endometrial hyperplasia cases having a personal or family history of cancer may belong to HNPCC, and that in these cases, abnormality of the mismatch repair system may be an early event in endometrial carcinogenesis.

Frequent hypermethylation of MLH1 promoter in normal endometrium of patients with endometrial cancers.

Silencing of the MLH1 gene by promoter hypermethylation is the mechanism underlying the microsatellite instability (MSI) phenotype in endometrial cancers. However, the profile of CpG methylation in a wide range of MLH1 promoters in endometrial cancers and in the normal endometrium is largely unknown. The present study investigates the region 700 bp upstream of MLH1 covering 48 CpG sites using bisulfite sequencing. Methylation status was classified as full (over 80% of CpGs are methylated), partial (10-80%) or nonmethylation (less than 10%). Of 56 endometrioid endometrial cancers, 16 (29%) were fully methylated, 14 (25%) were partially methylated and 26 (46%) were not methylated. Analyses of MLH1 by immunohistochemical means and of MSI revealed that the degree, rather than region-specific methylation of CpG islands is critical for decreased MLH1 expression and the MSI phenotype. Among 12 patients with methylated cancers, five (42%) patients contained methylated promoters in their normal endometria with profiles similar to those of cancer lesions, and these were associated with the MSI phenotype. In contrast, only one of 31 (3%) normal endometria from patients without endometrial malignancies harbored methylated promoters. These findings suggest that hypermethylation of the MLH1 promoter is frequent in the histologically normal endometrium adjacent to cancers, supporting the notion that hypermethylation of mismatch repair genes is the initial step that triggers various genetic events in endometrial carcinogenesis.

Mutational analysis of the PTEN gene in endometrial carcinoma and hyperplasia.

To determine the potential role of PTEN in the process of endometrial carcinogenesis, we examined a series of endometrial carcinoma and hyperplasia of the uterine corpus for the presence of a PTEN mutation. The entire coding region of the gene was screened for the presence of mutations by single-strand conformation polymorphism analysis, and mutations were confirmed by sequencing. We detected mutations in 14 of 57 endometrial carcinomas (13 of 50 endometrioid adenocarcinomas and 1 of 7 nonendometrioid adenocarcinomas) and 7 of 73 endometrial hyperplasias (1 of 24 simple hyperplasias without atypia, none of 16 complex hyperplasias without atypia, and 6 of 33 complex hyperplasias with atypia). Most (88%) mutations were clustered in exons 5, 7, and 8. Of the 24 mutations detected in 21 cases, 12 were frameshifts, 9 were nonsense, 2 were missense, and 1 was a silent mutation. Patients with a PTEN mutation had a better prognosis than those with no PTEN mutation. The presence of PTEN mutations in hyperplasia suggests that PTEN inactivation may occur as an initiating event in endometrial carcinogenesis and is involved in the development of cytologic atypia in hyperplasia.

MLH1 and MSH2 protein expression as a pre-screening marker in hereditary and non-hereditary endometrial hyperplasia and cancer.

The predictive value of MLH1 or MSH2 protein expression for the presence of truncating germline mutations was examined in benign and (pre)malignant endometrial samples from 3 patient groups: (I) 10 endometrial cancer patients from hereditary non-polyposis colorectal cancer (HNPCC) families with (n = 6) or without (n = 4) a known germline mutation; (II) 15 women from HNPCC families with (n = 7) or without (n = 8) a known germline mutation, who underwent endometrial sampling for non-malignant reasons; (III) 38 endometrial cancer patients <50 years of age, without HNPCC family history. Immunostaining for MLH1 and MSH2 was performed on paraffin-embedded sections. In group III, tumor DNA was examined for microsatellite instability (MSI) and MLH1, MSH2 and MSH6 mutation analysis was carried out. In 6/6 MLH1/MSH2 mutation carriers with endometrial cancer (group I), concordance was found between protein loss in the tumor and the corresponding mutation. In 3 MLH1 mutation carriers, MLH1 protein loss was also observed in concurrent endometrial hyperplasia. In group II, no protein loss was detected in normal endometrial tissue samples; in 3/4 patients with endometrial hyperplasia, MLH1/MSH2 protein loss was observed. In group III, protein loss was detected in 12/38 patients (9 MLH1, 3 MSH2), while in 3/11 patients with concurrent endometrial hyperplasia protein loss was also observed in the hyperplasia. MSI analysis in group III revealed 26 MSI-low and 12 MSI-high tumors. Mutation analysis in 28/38 patients showed only 1 missense MSH6 and no MLH1 or MSH2 germline mutations. In group III, loss of MLH1/MSH2 protein expression was not related to the presence of MSI or MLH1/MSH2 germline mutations. In conclusion, MLH1 or MSH2 protein loss in HNPCC-related endometrial neoplasia is strongly related to corresponding germline mutations. This relation was not clearly present in young sporadic endometrial cancer patients. Immunohistochemical pre-screening of the MLH1 and MSH2 proteins in endometrial hyperplasia or cancer can thus be helpful in HNPCC families. Frequent loss of MLH1 or MSH2 protein in endometrial hyperplasia indicates that this loss is an early event in endometrial carcinogenesis.

P53 nuclear accumulation in preneoplastic lesions and stage I uterine endometrioid adenocarcinoma

A total of 118 endometrial neoplastic and preneoplastic lesions comprising 43 uterine endometrioid adenocarcinoma at stage I, 40 complex (adenomatous) hyperplasias and 35 atypical hyperplasias were examined for p53 nuclear accumulation to assess the incidence of p53 alterations in infiltrating carcinomas and to verify if p53 aberrations may allow the identification of a subset of premalignant cases with high risk of progression. No specific immunostaining was observed in the cases of complex hyperplasia without atypias. One (3%) of 35 atypical hyperplasias showed focal areas of p53 immuno-reactivity. The overall frequency of p53 overexpression in endometrial carcinomas was 54%. The distribution of cases with nuclear accumulation of p53 was significantly different (p=0.01) in tumours with different degree of invasiveness. In addition, p53 nuclear accumulation was observed more often in tumours with moderate (G2) or poor differentiation (G3) (p=0.03). Our data indicate that p53 aberrations are not early events in endometrial carcinogenesis and may be related with tumour progression and aggressiveness.

Molecular events in endometrial carcinogenesis

Progress in understanding the molecular basis of carcinogenesis will undoubtedly have a major impact in gynaecologic oncology. Although the details of endometrial carcinogenesis are not as yet unravelled, important information has been accumulated in recent years. The identification of DNA repair genes responsible for the hereditary non-polyposis colorectal cancer (HNPCC) and Lynch II syndromes was an important step forward from a basic science perspective and should yield major benefits for individuals in these families. A number of steps in sporadic endometrial carcinogenesis have now been identified and it is possible to begin to construct a speculative pathway of multistep carcinogenesis. A single frequent genetic event in endometrial cancer has not yet been documented but several genetic alterations involving p53 mutation, Ki-ras mutation, loss of heterozygosity (LOH) at various chromosomal loci and amplification/overexpression of oncogenes such as c-myc and c-neu have been described. It is likely that as yet unknown and more frequent genetic alterations will eventually be identified. This information should be available during the next decade and as a result we can anticipate exciting progress in strategies for prevention, screening and therapy.