European Journal of Obstetrics & Gynecology and Reproductive Biology 120 (2005) 107–114 www.elsevier.com/locate/ejogrb
Genetic imbalances in endometrial hyperplasia and endometrioid carcinoma detected by comparative genomic hybridization Hamza M. Muslumanoglua, U. Onerb, S. Ozalpc, M.F. Acikalinb, O.T. Yalcinc, M. Ozdemira, S. Artana,* a
Department of Medical Genetics, Osmangazi University Medical Faculty, 26480 Eskisehir, Turkey b Department of Pathology, Osmangazi University Medical Faculty, Eskisehir, Turkey c Department of Obstetrics and Gynecology, Osmangazi University Medical Faculty, Eskisehir, Turkey Received 12 February 2004; received in revised form 28 May 2004; accepted 5 August 2004
Abstract Objective: To evaluate the sequential genomic copy alterations related to the development of precursor lesions and endometrioid-type endometrial carcinomas, and its association with cellular atypia. Study design: Paraffin-embedded tissue specimens from 32 cases of endometrial hyperplasia, 15 of endometrial carcinoma, and 20 of normal endometrial tissue were retrospectively evaluated by the comparative genomic hybridization (CGH) technique. The average number of copy alterations (ANCA) index was used to define the incidence of genomic imbalances in each tissue group. Identified sequential genetic abnormalities were compared with the final histopathological diagnosis and the cellular atypia. Results: Detectable and consistent chromosomal imbalances were found in 13 hyperplasia and 9 carcinoma specimens. There was a significant correlation between ANCA value and degree of cellular atypia and tumor grade. While 1p36-pter, 20q deletions, and 4q overrepresentation were the most prevalent imbalances detected in both complex hyperplasia and complex atypical hyperplasia, 17q22-qter deletion and amplification of 2p34 were only seen in hyperplasia with atypical cells. Overrepresentations of chromosomes 8q, 1q, and 3q are the most frequent aberrations in endometrial carcinomas, but were absent from all the precursor lesions except one. Underrepresentations of chromosomes 1p36-pter and 10q are the other commonly seen aberrations in carcinomas, the latter being more frequent in moderately differentiated than in poorly differentiated lesions. Conclusions: Different patterns of chromosomal aberrations are seen in precursor lesions than in endometrial carcinomas, except for the loss of 1p36-pter. The presence of 1p deletion in both endometrial hyperplasia and cancer specimens suggests that this is an early event in the development of carcinoma. These results support a stepwise mode of tumorigenesis with accumulation of a series of genomic copy alterations in endometrial carcinogenesis. # 2004 Elsevier Ireland Ltd. All rights reserved. Keywords: Genomic copy aberrations; Endometrial hyperplasia; Endometrial carcinoma; Comparative genomic aberrations; Endometrial carcinogenesis
1. Introduction Endometrial carcinoma is the most common malignancy of the female genital tract. In the past two decades, clinicopathological, immunohistochemical, and molecular genetic studies have provided new data, allowing for the development of a dualistic model of endometrial carcinogenesis. In this model, atypical hyperplasia is recognized as * Corresponding author. Tel.: +90 222 239 3771; fax: +90 222 239 2986. E-mail address:
[email protected] (S. Artan).
the precursor for the endometrioid type of endometrial carcinoma, and endometrial intraepithelial carcinoma, as the precursor for serous carcinoma [1]. On the other hand, Mutter et al. [2] have proposed that endometrial proliferations be broadly divided into monoclonal and polyclonal subgroups on the basis of the molecular genetic analysis of clonality. Polyclonal proliferations are designated ‘hyperplasia’, and monoclonal proliferations, ‘endometrial intraepithelial neoplasia’ (EIN) and have an elevated risk of progression to carcinoma. However, because clonality cannot be routinely performed on diagnostic specimens,
0301-2115/$ – see front matter # 2004 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.ejogrb.2004.08.015
108
H.M. Muslumanoglu et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 120 (2005) 107–114
this classification system is not currently recommended for use in clinical practice [1]. Many aspects of cell behavior are known to be affected by multiple genetic alterations in most cancers. These abnormal behaviors, including functional overexpression of protooncogenes, inactivation of tumor suppressor genes, and inefficiency of the mismatch repair system, could result in the dysregulation of cell growth. It is well documented that progression of many tumors requires the accumulation of genetic aberrations; this is known as the multistage theory of carcinogenesis [3,4]. As in other neoplastic diseases, the involvement of both genetic and environmental factors in carcinogenesis has been suggested in the case of endometrial carcinoma. Although the different types of endometrial hyperplasia differ in malignant potential, which is mostly related to their cellular atypia, the genetic events involved in the multistep transformation processes from normal endometrial tissue to endometrial hyperplasia and then to endometrioid-type endometrial carcinoma have not yet been clearly defined. Previous studies have demonstrated the alterations of K-ras, p53, DCC, c-erbB-2, and mismatch repair gene mutations in endometrial cancers [5–9]. Inactivation of tumor suppressor gene p53, which is obtained in 10–20% of cases of advanced-stage endometrial cancer, has not been detected in endometrial hyperplasia [10,11]. Significant allelic losses have been reported on several chromosome arms, such as 1p, 3p, 8p, 9p, 9q, 10q, 11q, 14q, 16q, 17p, and 18q, in endometrial cancer [12–14]. Allelic losses in several regions of chromosome 8p have frequently been detected in various tumors, including endometrial cancer, and the suspicion that chromosome 8p contains tumor suppressor genes associated with tumor progression has been entertained. This study was designed to determine genetic aberrations in endometrial hyperplasia, carcinoma, and normal tissue specimens retrospectively by the comparative genomic hybridization (CGH) technique in an attempt to define the genomic copy aberrations involved in the transformation processes from normal endometrium to invasive carcinoma.
2. Materials and methods Genomic DNA was extracted from 67 formalin-fixed, paraffin-embedded tissue specimens held in the archives of the Department of Pathology. The tissue specimens included endometrial hyperplasia in 32 cases and endometrioid-type endometrial carcinoma in 15; 20 specimens were of normal endometrial tissue, and these served as the control group. Four serial sections were obtained from each specimen, and the first and last sections were stained with hematoxylin and eosin stain and then reassessed by two authors of this paper who are pathologists and who were both blinded to the previous clinicopathological findings. The final histopathological diagnosis reached in each specimen, including any
cellular atypia and the grade of the lesions, was classified according to the FIGO criteria defined in 1988 [15]. The remaining two sections from each specimen were used for CGH analysis. The preparation of high-molecular-weight DNA from 15-mm endometrial tissue sections and from the peripheral blood leukocytes and the CGH analysis were performed as described previously [16]. Briefly, a section was collected from each formalin-fixed and paraffinembedded tissue specimen into an Eppendorf tube, where it was resuspended in 100 ml DNA extraction buffer (100 mM/l NaCl, 100 mM/l Tris–HCl, pH 7.6, 25 mM/l EDTA, 0.5% SDS) and 2 ml proteinase K (20 mg/ml) and incubated at 50 8C for 24 h. At the end of this period, the digested samples were then incubated at 95 8C for 10 min to inactivate the proteinase K. High-molecular-weight DNA was then extracted with phenol–chloroform and precipitated with ethanol. The reference DNA used in CGH analysis was isolated from peripheral blood of a karyotypically normal male. High-molecular-weight DNA from all lymphocytes was extracted using standard methods. The degenerate oligonucleotide-primed polymerase chain reaction (DOP-PCR) was used for uniform amplification of the DNA using the primer 6MW (50 -CCGACTCGAGNNNNNNATGTGG-30 ). Tumor and reference DNA probes were generated in a secondary labeling DOP-PCR reaction by incorporation of Spectrum Green (Vysis Inc.) and Spectrum Red (Vysis Inc.) directly conjugated nucleotides, respectively. After PCR labeling, DNA size was reduced to below 500 bp by means of a controlled deoxyribonuclease I digestion reaction that was allowed to proceed at room temperature for 10 min before being stopped by incubation at 95 8C for 2–3 min. Diploid metaphase spreads were obtained from peripheral blood lymphocyte cultures of normal male donors. Slide preparation was optimized to minimize residual cytoplasm and generate well-spread metaphases, as described previously [17]. The test and reference DNA samples, labeled in different colors, were precipitated together in the presence of 40 mg unlabeled Cot1DNA (Gibco BRL, Life Technologies Ltd., UK), resuspended in 12 ml hybridization mixture (50% formamide/2 SSC) denatured at 70 8C for 10 min, and preannealed for 1 h at 37 8C. The probe was hybridized to a normal denatured metaphase spread for 3 days, followed by stringency washes adapted from a standard fluorescence in situ hybridization protocol. Image acquisition and processing were performed by using MacProbe version 4.11 software (PSI system) in CGH analysis. Mean ratio profiles were determined from the analysis of at least 10 high-quality metaphase spreads. The lower and upper threshold values used to distinguish the genomic imbalances of the average ratio profiles were 0.75 and 1.20: shifts in the mean ratio values above 1.20 and below 0.75 were regarded as gains and deletions of chromosomal material, respectively. The average numbers of copy alterations (ANCA) were calculated as described by Kiechle et al. [18].
H.M. Muslumanoglu et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 120 (2005) 107–114
3. Results The final histopathological diagnoses, including cellular atypia and grade of the lesions and all genomic alterations identified by CGH in 67 tissue specimens (20 of normal endometrial tissues, 32 of endometrial hyperplasia, and 15 of endometrioid-type endometrial carcinoma), are summarized in Table 1. Fourteen (43.7%) of the endometrial hyperplasias were simple, nine (28.1%) were complex, and nine (28.1%) were both atypical and complex. Nine (60.0%) of the invasive carcinomas were grade II lesions, while five (33.3%) were grade III tumors and there was only one (6.6%) grade I tumor. CGH analysis detected loss of the 1p32-p35 region as a genomic abnormality in only 1 (5.0%) of the 20 normal endometrial specimens, which was taken from the proliferative endometrium of a 45-year-old woman. In contrast, 13 (40.6%) of the 32 hyperplastic endometrium specimens had detectable imbalances of the CGH profile: 3 (21.4%) of 14 simple hyperplasia, 4 (44.4%) of 9 complex hyperplasia, and 6 (66.7%) of the 9 complex atypical hyperplasia specimens. The incidence of genomic imbalances and the ANCA values were closely related to the histopathological subtype and the cellular atypia of the hyperplastic lesions. The ANCA values were observed to increase with increasing cellular atypia of the lesions, with 0.21, 1.56, and 2.78 for simple, complex, and complex atypical hyperplasias, respectively. While investigation of the simple hyperplasia samples revealed a total of three random genomic aberrations, the
109
chromosomal imbalances detected in complex hyperplasia with or without cellular atypia were recurrent aberrations. The most common recurrent aberrations were losses on 1p36-pter and 20q13-qter and gains on 4q24-q26. The loss on 16p was detected only in complex hyperplasia without atypia (3/9), whereas the loss on 17q22-qter and the gain on 2pter-p34 were found only in complex atypical hyperplasia (3/9) (Fig. 1a). Nine (60.0%) of the 15 endometrioid-type endometrial carcinomas had abnormally imbalanced CGH profiles. Endometrial carcinoma samples revealed a total of 52 genomic copy alterations. The mean ANCA value was 3.46 in endometrioid carcinomas. The 8q aberration was found in seven (46.7%) specimens and was thus the most common abnormality in endometrioid carcinomas. Both decreasing and increasing genomic copy alterations had similar frequencies in endometrioid carcinomas. The other recurrent chromosomal imbalances in the endometrioid carcinomas were losses on 1p36 and 10q, losses on 9q and 14q, and the loss on 11q, which were detected in five (33.3%), four (26.7%), and three (20.0%) specimens, respectively, and gains on 1q, 3q26, and 4p, which were observed in five (33.3%), four (26.7%), and three (20.0%) specimens, respectively (Fig. 1b). When the unbalanced CGH profiles and ANCA values were assessed with reference to the grade of endometrioid carcinoma, 5 (50.0%) of the 10 grade I or II and 4 (80.0%) of the 5 grade III tumors revealed genomic copy alterations. It was clear that both the incidence of an unbalanced CGH profile and the ANCA values tended to parallel the histopathological
Table 1 Histopathological data and chromosomal aberrations detected by CGH in endometrial specimens Case no.
Histopathological diagnosis
CGH analyses Deletion
Gain
18 32 33 34 37 38 39 43 44 45 48 49 50 52 53 55 56 58 61 62 64 66 67
Normal endometrium Simple hyperplasia Simple hyperplasia Simple hyperplasia Complex hyperplasia Complex hyperplasia Complex hyperplasia Complex hyperplasia Complex atypical hyperplasia Complex atypical hyperplasia Complex atypical hyperplasia Complex atypical hyperplasia Complex atypical hyperplasia Complex atypical hyperplasia EEC (G II) EEC (G II) EEC (G II) EEC (G II) EEC (G II) EEC (G III) EEC (G III) EEC (G III) EEC (G III)
1p32-35 17p – 22q14-qter 1pter-p36, 16p12, 20q13-qter 1pter-p36, 16p13, 20q13-qter 1pter-p36, 16p12 20q13-qter 1pter-p36, 17q22-qter, 20q13-qter 1pter-p36, 17q22-qter, 20q13-qter 9q34-qter 1pter-p36, 17q22-qter, 10q 1pter-p36 1pter-p36, 9q34-qter, 20q13-qter 10q 1pter-p36, 10q, 11q 10q, 11q 1pter-p36, 9q, 13q, 17q 1pter-p36, 9q, 10q 13q 10q, 11q 1pter-p36, 9q, 13q 1pter-p36, 2p, 9q, 13q, 17q, 20q
– – 8q – 4q24-q26, 6q16-q18 4q24-q26 4q24-q26, 6q16-q18
EEC: endometrioid endometrial carcinoma.
2pter-p34, 4q24-q26 2pter-p34, 4q24-q26 4q24-q26, 3q26, 6q16-q18 2pter-p34, 1q, 8q 18q 1q, 4p, 2q, 1q, 1q, 1q, 1q,
4p, 8q 5p, 8q 4p, 8q, 18q 3q26, 8q 3q26, 6q, 8q 2q, 3q26, 8q, 14q 3q26, 8q, 14q
110
H.M. Muslumanoglu et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 120 (2005) 107–114
Fig. 1. (a) CGH profiles of endometrial hyperplasia samples, showing gains (green) and losses (red) in individual chromosomes. (b) CGH profiles of endometrioid carcinomas, showing gains (green) and losses (red) in individual chromosomes. Thick bar, amplified chromosomal region. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of the article.)
grades. The ANCA value of grades I and II tumors was 2.4, whereas the corresponding value was 5.6 in grade III tumors. Gains on chromosomes 8q, 1q, and 3q26, and the different deleted chromosomal regions, were seen at higher frequencies in grade III endometrioid carcinomas.
4. Comment One of the fundamental features of cancer is tumor clonality, which is defined as the development of a tumor from a single cell that has begun to proliferate in an abnormal manner. Subsequently, any cell of this original
H.M. Muslumanoglu et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 120 (2005) 107–114
clone can acquire additional genetic alterations giving rise to all the characteristics of a cancer cell. Therefore, the development of cancer is accepted as a multistep process in which normal cells gradually progress to malignancy. Models for this concept have been proposed for the genetic alterations associated with tumor initiation and progression in some malignant human tumor types, including endometrial carcinoma. The malignant potential of endometrial hyperplasia was thought earlier to be related to its cellular atypia in most cases, but the genetic abnormalities involved in the multistep transformation processes leading from normal endometrial tissue to endometrial hyperplasia or invasive carcinoma have not yet been clearly defined. In an attempt to determine the genomic imbalances during the initiation and progression of endometrioid-type endometrial carcinoma, paraffin-embedded tissue specimens of endometrial hyperplasias with and without cellular atypia, invasive carcinomas with different grades, and normal endometrial tissues were screened by the genome-wide screening molecular cytogenetic method known as CGH. The CGH analysis performed in this study detected a loss of the 1p32-p35 region as a genomic abnormality in only 1 (5.0%) of the 20 normal endometrial specimens with a negligible risk of invasive carcinoma. However, long-term follow-up is needed for this patient because of the aberration involving chromosome 1p. In contrast, 40.6% of the endometrial hyperplasias had detectable DNA copy changes. Genomic imbalances were detected in 21.4% of simple, 44.4% of complex, and 66.7% of complex atypical hyperplasia, with risks of invasive carcinoma of 1%, 3%, and 29%, respectively [19]. Although the prevalence of chromosomal imbalances in simple hyperplasia was low and the types of aberrations were random, the aberration frequency and the recurrence rate of imbalances increased with increasing architectural complexity and cellular atypia. This observation was also confirmed by the higher ANCA values observed in architecturally complex specimens with cellular atypia. The ANCA values gradually increased from 0.21 in simple hyperplasia through 1.56 in complex hyperplasia to 2.78 in complex atypical hyperplasia. These findings suggested that the increased rate and higher grade of genomic imbalance reflected in higher ANCA values might be closely related to the degree of architectural complexity and cellular atypia. Accumulation of the defined genomic DNA aberrations in these chromosomal regions of cell clones could be involved in the initiation of precursor lesions, which might progress to invasive carcinoma as a result of malignant transformation. This suggestion is also supported by the findings of earlier studies [18,20–22]. Although losses of 1pter-p36 and 20q13-qter and gains of 4q24-q26 were the most common aberrations found in complex lesions with or without atypia, 16p deletion was observed only in those without atypia and loss of 17q22-qter and gain of 2pter-p34 was seen only in complex atypical lesions. It is suggested that the presence of these recurrent
111
and specific genomic alterations only in complex lesions reflects the molecular changes involved in the generation and progression of these lesions. The 1p36 region contains several putative cancer-related genes, such as the CDC2L1, PRDM2, and E2F2 genes. The cell division cycle2-like 1 (CDC2L1) gene encodes wellconserved protein kinase p58, which is known to be essential for eukaryotic cell cycle control. This cell division controlrelated gene has been shown to be frequently deleted or altered in neuroblastoma with amplified N-myc genes, ductal breast carcinoma, endocrine neoplasia, and malignant melanoma [18]. PRDM2, located at 1p36, is a tumor suppressor gene that is a member of a nuclear histone protein methyltransferase superfamily. The gene encodes a zinc finger protein that can bind to retinoblastoma protein and estrogen receptor. The RIZ1 tumor suppressor gene is a member of a methyltransferase superfamily, and RIZ1 inactivation is commonly found in many types of human cancers. The other gene located at 1p36 is the E2F2 gene. The protein encoded by this gene is a member of the E2F family of transcription factors, which act as transcriptional activators that are important for progression through the G1/ S transition. This study revealed deletion of the chromosome 1pter-p36 region in both complex and atypical complex lesions, and it was also the most common aberration seen in invasive endometrial carcinoma. This abnormality has also been reported by Kiechle et al. [18] as a consistent aberration in complex hyperplasia. These findings suggest that aberrations of the gene(s) located on chromosome 1pterp36 region might have significant roles in the development of lesions that involve a greater risk of invasive cancer. This study shows that endometrial complex hyperplasia without atypia is also characterized by deletion of chromosome 16p, which encodes epithelial membrane protein 2; this protein is involved in cell proliferation and cell-to-cell interactions and is one of the ubiquitinconjugating enzymes that are involved in such essential cellular processes as DNA repair, cell cycle control, and stress responses [23]. The deletion of this gene locus might be involved in uncontrolled proliferation of the endometrial cells. The other common genomic alteration obtained by this study in endometrial complex hyperplasia with or without atypia cells was the loss of the chromosome 20q13-qter region, in which one of the other members of the ubiquitinconjugating enzyme family is also located. In addition, CSE1L/CAS protein, which is encoded by the CAS gene located in the 20q13 region, is a Ran-binding protein that has an important role at the mitotic spindle checkpoint, which is important for genomic stability during cell division. Deregulation of proliferation and apoptosis, which is well known to contribute to neoplastic transformation and uncontrolled growth, has been reported to occur in some neoplasms as a result of disturbance of the mitotic spindle checkpoint [24]. The deletion of 20q13-qter could, then, be involved in the progressive deregulation of proliferation and
112
H.M. Muslumanoglu et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 120 (2005) 107–114
apoptosis that might cause endometrial hyperplasia to progress to invasive carcinoma [25,26]. Although the genomic aberrations detected in complex hyperplasia in this study were similar to those found by Kiechle et al. [18], Baloglu et al. [22] reported completely different aberrations. Baloglu’s group obtained chromosome 1 amplification most frequently in atypical hyperplasia and invasive carcinoma, but no chromosome 4q imbalance was determined in their specimens. In contrast, the gain on chromosome 1q was one of the most frequently seen aberrations in invasive carcinoma in this study, as reported previously [12]. When compared with results obtained in endometrial hyperplasia, the results of this study reveal that endometrioid carcinoma is characterized by totally different genomic aberrations, except for loss of the 1p36-pter region. Further differences are the more frequently increased fluorescence intensities, the accumulation of genetic aberrations, and the frequency of chromosomal imbalances, which were all more significant in endometrioid carcinomas. The corresponding ANCA values were also higher in invasive lesions and were observed to increase with increasing grade of the carcinoma. The findings support the concept of a significant correlation between the accumulation of genetic aberrations and the degree of cellular atypia and grade of endometrioid carcinoma, which probably develop in a multistep process with the acquisition of genetic changes. Loss of the chromosome 1p36 region was a common aberration in endometrioid carcinomas in this study. The presence of this aberration even in low-grade tumors suggests that the mutations in the gene(s) mapped to this chromosomal region might be involved in tumor development and progression. Detection of deletion of the chromosome 1p36-pter region in specimens of both complex hyperplasia and endometrioid carcinoma indicates the importance of gene(s) located in this region in tumor progression. The prevalence of this alteration was higher in atypical hyperplasia, in which there is a risk of tumor progression. We suggest that loss of 1p36-pter might be an early event in hyperplasia and that tumor development and progression might be triggered through the additional genomic changes. In our series, combined losses of chromosomes 1p36-pter and 9q were detected in four carcinoma samples. PTEN is a multifunctional tumor suppressor gene that is involved in the induction of apoptosis, blocking the cell cycle at the G1 stage. Several groups have reported finding mutational inactivation of the tumor suppressor gene PTEN
in 33–55% of both low- and high-grade endometrioid carcinomas [27,28]. The loss of PTEN expression in endometrioid carcinomas and precancerous lesions has been suggested to have a ‘gatekeeper’ role [29,30]. The frequent presence of 10q deletion found in low- and highgrade endometrioid-type carcinomas in this study strongly suggests involvement of the gene encoded by this locus, possibly PTEN, in development and progression of endometrioid carcinoma. In a similar way to earlier studies, this study reveals that gains of chromosome 1q and 8q are also frequent aberrations in invasive carcinomas, especially in high-grade lesions [12,31,32]. These findings suggest that the genes located on these chromosomal bands might be important in tumor progression. Minimal overlapping regions of the gains, which are indicative of the presence of oncogenes, were mapped to 1q22-qter and 8q24-qter. Chromosomal gains on 1q21-22 and 1q23-24 have been reported as frequently seen aberrations in various cancers, such as bladder, hepatocellular, and breast carcinoma [33], and it has been suggested that gains of DNA in these regions are important for cancer development and progression. The Rit1 gene mapped to the 1q21.3 locus is a member of the Ras superfamily. Rit1 protein, with a similar molecular structure to the Ras protein, is a putative oncogene that might be important in the regulation of signaling pathways [33]. Protein expression studies have shown significant amplification of this protein especially in patients with short survival times. In contrast to some earlier studies, which revealed no significant association between overrepresentation of sequences on chromosome 8q22-23 and 8q24-qter and the aggressive clinical behavior of various cancers, the minimal overlapping region of chromosome 8q24-qter, a common aberration in high-grade endometrioid carcinomas, was the most frequently seen genetic abnormality in endometrial carcinoma specimens examined in this study [32,34]. Suehiro et al. [32] described a relationship between the detected copy number gains at 8q24-qter and lymph node metastasis. Moreover, significant overexpression of cytoplasmic protein tyrosine kinase, encoded by the PTK2/FAK gene located on the 8q24-qter region, has been reported in some high-grade carcinomas [35]. It is suggested that this putative oncogene is an important marker for carcinogenesis and tumor progression. Chromosome gain on 3q26 was seen only in poorly differentiated endometrioid carcinoma samples. This result suggests that increasing copy number on chromosome 3q26
Fig. 2. A model of the stepwise accumulation of genetic imbalances during tumorigenesis of endometrioid-type endometrial carcinomas.
H.M. Muslumanoglu et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 120 (2005) 107–114
might be involved in tumor progression. Genes such as TERC (telomerase RNA component) and PLD (phospholipase D) map to chromosome region 3q26, and links between these genes and tumor progression have been shown in previous studies [36–38]. It is well known that telomerase, a ribonucleoprotein polymerase that maintains telomere ends by addition of the telomere repeat, is normally repressed in postnatal somatic cells, resulting in progressive shortening of telomeres. hTERC is one of the main components of telomerase and is involved in telomerase reactivation. Previous studies showed that hTERC expression is regulated during carcinogenesis [37]. The PLD gene, which is located on 3q26, stimulates cell growth by repressing expression of the p21 gene, so that it is probably related to carcinogenesis [38].
5. A model of the genomic alterations associated with endometrial cancer progression According to the concept of clonal evolution, the development of a tumor is initiated by clonal expansion of a single cell carrying a mutation that leads to a growth advantage. Subsequently, any cell of this original clone can acquire additional genetic alterations, giving rise to more rapidly growing subclones. Tumors thus progress in a multistep process by way of cumulative acquisition of genetic changes. In the presence of endometrial lesions, a model of the most likely sequence of chromosomal and genetic alterations based on the genomic changes summarized in Fig. 2 is mostly in line with the model proposed by Kiechle et al. [18]. However, there are some differences between these algorithms, and the model suggested here therefore needs to be refined by further molecular studies conducted in larger study groups to determine chromosomal localizations and the genes located at these regions involved in tumor development and progression in the endometrium. Although CGH is one of the more robust technologies employed for determining alterations in gene dosage, the fact that it can detect only general regions of chromosomal alterations is an important limitation of it. A recently developed technology known as CGH-microarray possesses a higher resolution capability and can also localize the altered region in the chromosome and determine putative cancer-related genes. Therefore, the model suggested in this paper might be refined, and the nature of sequential genetic events in endometrioid-type endometrial carcinomas would be clarified by CGH-microarray and gene expression studies.
References [1] Kurman RJ, editor. Blaustein’s pathology of the female genital tract. 5th ed. New York: Springer; 2002.
113
[2] Mutter GL, Baak JPA. Crum CP, et al. Endometrial precancer diagnosis by histopathology, clonal analysis, and computerized morphometry. J Pathol 2000;190:462–9. [3] Fearon ER, Vogelstein B. A genetic model for colorectal tumorigenesis. Cell 1990;61:759–67. ¨ T, Basaran N, Kabukc¸uog˘ lu S, Minsin T. Prognostic [4] Ozalp S, Yalc¸ın O significance of deletion and over-expression of the p53 gene in epithelial ovarian cancer. Eur J Gynaecol Oncol 2000;21:282–6. [5] Goodfellow PJ, Buttin BM, Herzog TJ, et al. Prevalence of defective DNA, mismatch repair and MSH6 mutation in an unselected series of endometrial cancers. Proc Natl Acad Sci USA 2003;100: 5908–13. [6] Helou K, Walentinsson A, Beckmann B, et al. Analysis of genetic changes in rat endometrial carcinomas by means of comparative genomic hybridization. Cancer Genet Cytogenet 2001;127: 118–27. [7] Taskin M, Lallas TA, Barber HR, Shevchuk MM. Bcl-2 and p53 in endometrial adenocarcinoma. Mol Pathol 1997;10:728–34. [8] Ronnett BM, Burks RT, Cho KR, Hedrick L. DCC genetic alterations and expression in endometrial carcinoma. Mod Pathol 1997;10:38–46. [9] Elhafey AS, Papadimitriou JC, El-Hakim MS, El-Said AI, Ghannam BB, Silverberg SG. Computerized image analysis of p53 and proliferating cell nuclear antigen expression in benign, hyperplastic and malignant endometrium. Arch Pathol Lab Med 2001;125:872–9. [10] Enomoto T, Fujita M, Inoue M, et al. Alterations of the p53 tumor suppressor gene and its association with activation of the K-ras-2 protooncogene in premalignant and malignant lesions of the human uterine endometrium. Cancer Res 1993;53:1883–8. [11] Ozalp S, Yalcin OT, Tanir HM, Kabukcuoglu S, Erol G. p53 overexpression as a prognostic indicator in endometrial carcinoma. Eur J Gynaecol Oncol 2003;24:275–8. [12] Pere H, Tapper J, Wahlstrom T, Knuutila S, Butzow R. Distinct chromosome imbalances in uterine serous and endometrial carcinomas. Cancer Res 1998;58:892–5. [13] Muresu R, Siri MC, Cossu A, et al. Chromosomal abnormalities and microsatellite instability in sporadic endometrial cancer. Eur J Cancer 2002;38:1802–9. [14] Sirchia SM, Sironi E, Grati FR, et al. Losses of heterozygosity in endometrial adenocarcinomas: positive correlation with histopathological parameters. Cancer Genet Cytogenet 2000;121: 156–62. [15] Benedet JL, Hacker NF, Ngan HYS, editors. Staging classifications and clinical practice guidelines of gynecologic cancers by FIGO Committee on Gynecologic Oncology and IGCS Guidelines Committee. 2nd ed. Canada: Elsevier Science; 2000. [16] Arslantas A, Artan S, Oner U, et al. The importance of genomic copy number changes in the prognosis of glioblastoma multiforme. Neurosurg Rev 2004;27:58–64. [17] Henegariu O, Heerema NA, Wright LL, Bray-Ward P, Ward DC, Vance GH. Improvements in cytogenetic slide preparation: controlled chromosome spreading, chemical aging and gradual denaturing. Cytometry 2001;43:101–9. [18] Kiechle M, Hinrichs M, Jacobsen A, et al. Genetic imbalances in precursor lesions of endometrial cancer detected by comparative genomic hybridization. Am J Pathol 2000;156:1827–33. [19] Kurman RJ, Kaminski PF, Norris HJ. The behavior of endometrial hyperplasia. A long-term study of untreated hyperplasia in 170 patients. Cancer 1985;56:403–12. [20] Heselmeyer K, Schrock E, du Manoir S, et al. Gain of chromosome 3q defines the transition from severe dysplasia to invasive carcinoma of the uterine cervix. Proc Natl Acad Sci USA 1996;93:479–84. [21] Ried T, Heselmeyer-Haddad K, Blegen H, Schrock E, Auer G. Genomic changes defining the genesis progression and malignancy potential in solid human tumors: a phenotype/genotype correlation. Genes Chromosomes Cancer 1999;25:195–204. [22] Baloglu H, Cannizzaro LA, Jones J, Koss LG. Atypical endometrial hyperplasia shares genomic abnormalities with endometrioid carci-
114
[23]
[24]
[25]
[26]
[27] [28]
[29] [30]
H.M. Muslumanoglu et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 120 (2005) 107–114 noma by comparative genomic hybridization. Hum Pathol 2001;32:615–22. Ben-Porath I, Benvenisty N. Characterization of a tumor-associated gene, a member of a novel family of genes encoding glycoproteins. Gene 1996;183:69–75. Wellmann A, Flemming P, Behrens P, et al. High expression of the proliferation and apoptosis associated CSE1L/CAS gene in hepatitis and liver neoplasms: correlation with tumor progression. Int J Mol Med 2001;7:489–94. Peiro G, Diebold J, Baretton GB, Kimmig R, Lohrs U. Cellular apoptosis susceptibility gene expression in endometrial carcinoma: correlation with Bcl-2 Bax and caspase-3 expression at outcome. Int J Gynecol Pathol 2001;20:359–67. Brinkmann U. CAS, the human homologue of the blast chromosome segregation gene CSE1, in proliferation, apoptosis and cancer. Am J Hum Genet 1998;62:509–13. Kong D, Suzuki A, Zou TT, et al. PTEN is frequently mutated in primary endometrial carcinomas. Nat Genet 1997;17:143–4. Hirai Y, Tanaka N, Furuta R, et al. Somatic mutations of the PTEN/ MMAC1 gene associated with frequent chromosomal loss detected using comparative genomic hybridization in endometrial cancer. Gynecol Oncol 2001;83:81–8. Ali IU. Gatekeeper for endometrium: the PTEN tumor suppressor gene. J Natl Cancer Inst 2000;92:861–3. Maxwell GL, Risinger JI, Gumbs C, et al. Mutation of the PTEN tumor suppressor gene in endometrial hyperplasias. Cancer Res 1998;58:2500–3.
[31] Suzuki A, Fukushige S, Nagase S, Ohuchi N, Satomi S, Horii A. Frequent gains on chromosome arms 1q and/or 8q in human endometrial cancer. Hum Genet 1997;100:629–36. [32] Suehiro Y, Umayahara K, Ogata H, et al. Genetic aberrations detected by comparative genomic hybridization to predict outcome in patients with endometrioid carcinoma. Genes Chromosomes Cancer 2000;29: 75–82. [33] Li JT, Liu W, Kuang ZH, et al. Amplification of RIT1 in hepatocellular carcinoma and its clinical significance. Ai Zheng 2003;22:695–9. [34] Wolter H, Gottfried HW, Mattfeldt T. Genetic changes in stage pT2N0 prostate cancer studied by comparative genomic hybridization. BJU Int 2002;89:310–6. [35] Moritake H, Sugimoto T, Kuroda H, et al. Newly established Askin tumor cell line and overexpression of focal adhesion kinase in Ewing sarcoma family of tumors cell lines. Cancer Genet Cytogenet 2003;146:102–9. [36] Heselmeyer-Haddad K, Janz V, Castle PE, et al. Detection of genomic amplification of the human telomerase gene (TERC) in cytologic specimens as a genetic test for the diagnosis of cervical dysplasia. Am J Pathol 2003;163:1405–16. [37] Guilleret I, Yan P, Guillou L, Braunschweig R, Coindre JM, Benhattar J. The human telomerase RNA gene (hTERC) is regulated during carcinogenesis but is not dependent on DNA methylation. Carcinogenesis 2002;23:2025–30. [38] Kwun HJ, Lee JH, Min do S, Jang KL. Transcriptional regression of cyclin-dependent kinase inhibitor p21 gene by phospholipase D1 and D2. FEBS Lett 2003;544:38–44.
