About the Authors Hidemichi Watari Hidemichi Watari is Clinical Associate Professor at the Department of Gynecology, Hokkaido University Hospital (Sapporo, Japan). His research interests focus on the mechanisms of treatment resistance (chemoresistance and radioresistance) in gynecologic malignancies, and the discovery of new therapeutic targets and/or biomarkers to predict therapeutic response.
Peixin Dong Peixin Dong is an Assistant Professor of Women’s Health Educational System at Hokkaido University. He studies the molecular mechanisms of cancer metastasis, especially the roles of mutant p53 gain of function in endometrial cancer progression and the regulatory effects of miRNAs on epithelial–mesenchymal transition.
Takashi Mitamura Takashi Mitamura is part of the medical staff at the Department of Gynecology, Hokkaido University Hospital. His research interests focus on the mechanisms of chemoresistance in ovarian cancer and the discovery of new therapeutic targets and/or biomarkers to predict therapeutic response.
Hiromasa Fujita Hiromasa Fujita is a Cytopathologist in Hokkaido Cancer Society (Sapporo, Japan). His research focuses on clarifying whether endometrial cytology may be helpful for the diagnosis of early endometrial carcinomas with serous features, including endometrial intraepithelial carcinoma or precancerous lesions (p53 signatures).
Noriaki Sakuragi Noriaki Sakuragi is Head of the Department of Gynecology at Hokkaido University Hospital, and Chair of the Department of Gynecology at Hokkaido University Graduate School of Medicine. His main areas of clinical and research interest focus on gynecological cancers. He and his colleagues have developed and published their technique of an anatomically based nerve-sparing radical hysterectomy for invasive cervical cancer. His expertise also involves endometrial cancer. For reprint orders, please contact:
[email protected]
34
34
© 2014 Future Medicine www.futuremedicine.com
Chapter
3 Genetic model of endometrial carcinogenesis
Molecular genetic alterations in type I tumors36 Molecular genetic alterations in type II tumors39 New concept of molecular genetic model of serous carcinoma40 Molecular genetic alterations in carcinosarcoma40 Noncoding RNA for endometrial carcinogenesis41 Long noncoding RNAs
41
Cancer stem cells for endometrial carcinogenesis42 Conclusion42
doi:10.2217/FMEB2013.14.4
© 2014 Future Medicine
Hidemichi Watari, Peixin Dong, Takashi Mitamura, Hiromasa Fujita & Noriaki Sakuragi Endometrial cancer (EC) is the most common cancer of the female genital tract. EC frequently occurs in peri- and postmenopausal women. From a clinical point of view, EC is classified into two different types, types I and II [1]. Type I tumors are low-grade and estrogen-related endometrioid endometrial carcinomas that usually develop in perimenopausal women and coexist or are preceded by complex and atypical endometrial hyperplasia. By contrast, type II tumors are aggressive non-endometrioid endometrial carcinomas (serous and clear cell carcinomas) that occur in older women, and are not related to estrogen stimulation. Occasionally, type II carcinomas may arise in association with ‘serous endometrial intraepithelial carcinoma’, either from atrophic endometrium or endometrial polyps. It has been demonstrated that molecular genetic alterations involved in the development of type I carcinomas are different from those of type II carcinomas (Table 3.1) [2,3]. In this chapter, we describe the recent genetic models of endometrial carcinogenesis (Table 3.2 & Figure 3.1).
35
Watari, Dong, Mitamura, Fujita & Sakuragi Molecular genetic alterations in type I tumors Type I tumors show microsatellite instability (MI) and mutations in PTEN, PIK3CA, KRAS and b-catenin genes. MI was originally identified in patients with the hereditary nonpolyposis colorectal carcinoma (HNPCC). However, MI was also found in sporadic colon cancers. Endometrial cancer (EC) is the second most common cancer found in patients with HNPCC. MI is observed more frequently in EC associated with HNPCC than in sporadic EC [4,5]. HNPCC patients show an inherited germline mutation in MLH1, MSH2, MSH6 and PMS2. MI is known to occur more frequently in type I tumors than in type II tumors. In sporadic EC, mismatch repair deficiency is mainly caused by epigenetic inactivation of MLH1 (promoter hypermethylation) [6]. The MI-associated mismatch repair deficiency leads to the accumulation of mutations in coding and noncoding DNA sequences, including shorttandem repeats, named microsatellites. Some small short-tandem repeats, such as mononucleotide repeats, are located within the coding sequence of some important genes (BAX, IGFIIR, hMSH3, hMSH6, MBD4, CHK-1, Caspase-5, ATR, ATM, BML, RAD-50, BCL-10 and Apaf-1), and they may be potential targets in the process of tumor progression of EC with MI [7,8]. The tumor-suppressor gene phosphatase and tensin homolog deleted on chromosome 10 (PTEN) is frequently abnormal in EC [9–11]. Somatic PTEN mutations are common in EC, and observed predominantly in type I tumors, occurring in 37–61% of cases, and lead to activation of the PI3K/AKT pathway. PTEN mutations are found in 60–86% of MI-positive type I tumors, but in only 24–35% of the MI-negative type I tumors. Identical PTEN mutations have been detected in hyperplasias coexisting with MI-positive type I tumors, suggesting that PTEN mutations are early events in type I tumor development. Since type I tumors with PTEN mutation have been shown to have genomic instability [12], poly(ADP-ribose) polymerase inhibitors might be active in some type I tumors [13]. Mutations in PI3CA lead to alteration of the PI3K/AKT pathway in EC [14–16]. PI3K is a heterodimeric enzyme consisting of a catalytic subunit (p110) and a regulatory subunit (p85). The Type I endometrial cancers (ECs) show micro PIK3CA gene, which is located on satellite instability and mutations in PTEN, chromosome 3q26.32, and encodes for the PIK3CA, KRAS and b-catenin genes. PTEN mutations are found in 37–61% of ECs. PIK3CA mutations occur p110a catalytic subunit of PI3K, has recently in 24–39% of ECs, and are frequently found with PTEN been reported to be mutated in EC. mutations. KRAS mutation has been reported in Mutations are frequently identified in 10–30% of ECs, and more frequently in type I tumors exon 9 and exon 20, and occur in exons 1–7 with microsatellite instability. Mutations in exon 3 of [17]. PIK3CA mutations occur in 24–39% of b-catenin are observed in 14–44% of ECs, which result in stabilization of the protein, change of intracellular EC, and are frequently found with PTEN localization, and participation in signal transduction mutations. PIK3CA mutations were shown and transcriptional activation.
36
www.futuremedicine.com
Genetic model of endometrial carcinogenesis Table 3.1. Comparison of clinicopathologic characteristics between type I and type II tumors. Characteristic
Type I
Type II
Age
Perimenopause
Postmenopause
Incidence
80%
20%
Risk factors
Unopposed estrogen, obesity and DM
Breast cancer history
Histologic type
Endometrioid
Serous, clear cell
Precursor
AEH
EmGD
Differentiation
Well to moderately differentiated
Poorly differentiated
Prognosis
Good
Poor
AEH: Atypical endometrial hyperplasia; DM: Diabetes mellitus; EmGD: Endometrial glandular dysplasia.
to be associated with poor prognostic factors. Although initially described in type I tumors, PI3KCA mutations also occur in type II tumors [18,19]. Furthermore, different gene-expression profiles in the PI3K–AKT signaling pathway serve to separate two subgroups of high-grade EC with distinct molecular alterations (PI3K–AKT pathway versus p53 alterations) that may play different roles in endometrial carcinogenesis [20]. Moreover, mutations in PIK3RI (p85a), the inhibitory subunit of PI3K, have been detected in 43% of type I tumors and 12% of type II tumors [21]. Among AKT targets, mTOR might be a new therapeutic target for EC, because mTOR inhibitors have recently been developed and introduced into clinical practice. mTOR inhibitors are expected to be active in EC with PTEN inactivation, because pharmacological inhibition of mTOR in PTEN+/- mice has shown reduced Table 3.2. Comparison of molecular alterations between type I and type II tumors. Molecular alteration
Type I (%)
Type II (%)
ER expression
30–70
