| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Molecular Oncology, Markers, Clinical Correlates |
Department of Pathology, The Norwegian Radium Hospital, Montebello, N-0310 Oslo, University of Oslo, Norway [B. D.]; Department of Pathology [I. G., J. K.] and Division of Gynecologic Oncology [W. H. G., G. B-B.], Sheba Medical Center, Tel-Hashomer 52621, Israel (affiliated with Sackler School of Medicine, Tel-Aviv University); and Department of Pharmacology, Faculty of Medicine, and Center for Pharmacy, Hebrew University, Jerusalem 91120, Israel [R. R.]
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Experimental Design: Sections from 61 primary ovarian carcinomas and metastatic lesions from 36 patients diagnosed with advanced-stage ovarian carcinoma [International Federation of Gynecologists and Obstetricians (FIGO) stages III-IV] were evaluated for expression of PEA3 using mRNA in situ hybridization. Patients were divided into long-term (n = 16) and short-term (n = 20) survivors.
Results: The mean values for disease-free survival and overall survival were 119 and 137 months for long-term survivors, as compared with 4 and 22 months for short-term survivors, respectively. Expression of PEA3 mRNA was detected in carcinoma cells and stromal cells in 56 of 61 lesions (92%) and 54 of 61 lesions (89%), respectively. Intense stromal expression was detected only in the vicinity of grade 2–3 tumors (P = 0.04). PEA3 expression in stromal cells showed a significant association with matrix metalloproteinase 2 mRNA expression in carcinoma cells (P = 0.022). PEA3 expression in carcinoma cells showed an association with mRNA expression of the ß1 integrin subunit in the same compartment (P = 0.039). It was also associated with mRNA expression of ß1 integrin subunit (P = 0.012), basic fibroblast growth factor (P = 0.036), and the matrix metalloproteinase inducer EMMPRIN (P = 0.038) in stromal cells. PEA3 mRNA was detected more often in both carcinoma and stromal cells in tumors of short-term survivors (P = 0.021 for stromal cells). In univariate survival analysis, PEA3 expression in stromal cells correlated with both shorter disease-free survival (P = 0.019) and overall survival (P = 0.029), whereas tumor cell expression predicted poor overall survival (P = 0.049). PEA3 mRNA expression in stromal cells emerged as an independent predictor of poor outcome in multivariate survival analysis, in which all molecules previously studied in this patient cohort were included (P = 0.015).
Conclusions: To the best of our knowledge, this is the first evidence associating PEA3 mRNA expression and poor survival in human epithelial malignancy. PEA3 is thus a novel prognostic marker in advanced-stage ovarian carcinoma. The association between PEA3 mRNA expression and the expression of the ß1 integrin subunit, basic fibroblast growth factor, and EMMPRIN, first documented in our patient cohort, points to the central role of this transcription factor in tumor progression in ovarian carcinoma.
|
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The Ets family of transcription factors is divided into subfamilies (Ets-1 and -2, ERG, GABP, PEA3, ELK, ELF, and PU1) based mainly on the sequence and location of the Ets domain, an 84-amino acid sequence present in all members of the family. Ets proteins bind to DNA sequences with the core motif C/A-GCA-A/T (3) . Ets transcription factors play a role in a variety of physiological and pathological processes, including embryogenesis, wound healing, and tumor progression (3, 4, 5) . This is largely due to their ability to activate the transcription of proteases, including urokinase-type plasminogen activator (6) and MMPs3 (7) , as well as that of TIMPs (7) and ß3 integrin (8) . The activation of proteolytic enzyme transcription is central to the metastatic process because of the role of these enzymes in both angiogenesis and tumor invasion.
The PEA3 subfamily includes PEA3, ER81, and ERM (9, 10, 11) . The human PEA3 homologue, E1AF, shares a 94% homology with PEA3 and has an identical Ets domain (9 , 12) . It binds to the enhancer elements of the adenovirus type 5 E1A gene (13) and has been shown to up-regulate MMP transcription (14) . Among epithelial malignancies, breast cancer has been the most extensively studied. PEA3 overexpression was reported in primary and metastatic lesions of mouse mammary carcinoma (15) , and expression of its dominant-negative form reduced tumor growth in this model (16) . PEA3 mRNA expression showed an association with HER2/neu expression, but not with estrogen receptor status, nuclear grade, or S-phase fraction, in clinical specimens of breast carcinoma (17) , and PEA3 protein expression predicted worse overall survival in this malignancy (18) . Conversely, PEA3 expression suppressed HER2/neu expression in human breast and ovarian carcinoma (19) and mediated apoptosis in the SKBr-3 breast carcinoma cell line (20) . In addition, PEA3 subfamily members were detected in benign breast epithelium and estrogen receptor-positive tumor cells, but not in their receptor-negative counterparts (21) . PEA3 expression was found in the majority of both clinical specimens and cell lines of lung and oral carcinoma (22, 23, 24) . Transfection with PEA3 resulted in enhanced motility and invasion in lung cancer cells (22) , whereas transfection of oral carcinoma cells with an antisense sequence resulted in inhibition of invasion and MMP expression (23) . As opposed to breast and lung tumors, PEA3 was reported to be absent in prostate carcinoma (25) . Although reported in ovarian carcinoma cell lines (26) , PEA3 expression in patient material and its role in the clinical course of ovarian carcinoma are unknown to date.
Ovarian cancer is the sixth most common cancer and the sixth most frequent cause of cancer death in women (4.4% of cases and 4.5% of deaths), and it is the leading cause of death from gynecologic cancer in women in industrialized countries (27)
. Despite the inclusion of new chemotherapeutic regimens, the mortality rate from ovarian carcinoma has remained essentially unchanged, mainly due to the late clinical presentation of this tumor. However, it is crucial to attempt to further characterize different prognostic groups of patients within each given stage by the use of molecular markers. Studying the same patient cohort in a series of investigations of metastasis-related gene expression, we have been able to highlight the role of mRNA expression of Ets-1 (4)
, MMPs and their inhibitor TIMP-2 (28)
, and the 
v integrin subunit (29)
as a predictive factor for disease aggressiveness in ovarian carcinoma. Conversely, mRNA expression of the MMP inducer EMMPRIN4
and the angiogenic factors bFGF, interleukin 8, and vascular endothelial growth factor (30)
did not influence the prognosis of patients in this cohort. The present study evaluates the expression and prognostic value of PEA3 mRNA in ovarian carcinoma and investigates the potential association between PEA3 expression and that of the previously studied molecules.
|
PATIENTS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Tumors.
Sixty-one formalin-fixed, paraffin-embedded blocks from the archives of the Department of Pathology at the Sheba Medical Center were included in the study. These consisted of 32 primary ovarian tumors and 29 metastatic lesions from 36 patients with advanced ovarian carcinoma. The distribution of the studied material according to biopsy site is shown in Table 1
. Sections from all tumors were reviewed by two observers (B. D. and J. K.) in consensus sessions to confirm the diagnosis, histological type, and tumor grade (I-III, corresponding to well, moderately, and poorly differentiated). Established criteria were used for the microscopic diagnosis and tumor classification (31)
. Tumor staging was established according to FIGO criteria.
|
The specificity of the probe was verified using a sense probe (Research Genetics). The sense probe sequence was as follows: 5'-CGG-CTC-AGC-TTC-TCA-TAA-TTC-A-3'.
A poly d(T)20 oligonucleotide probe (Research Genetics) was used to verify the integrity and lack of degradation of mRNA in each sample. cDNA probes were hyperbiotinylated. The stock dilution was diluted with probe diluent (Research Genetics) immediately before use.
mRNA ISH.
Tissue sections (4 µm thick) of formalin-fixed, paraffin-embedded specimens were mounted on ProbeOn Plus slides (Fisher Scientific, Pittsburgh, PA). Sectioning was performed in RNAase-free water. mRNA ISH was carried out by using the microprobe manual staining system (Fisher Scientific; Ref. 32
). Hybridization of the probes was carried out as described previously (28)
. A positive enzymatic reaction in this assay stained dark blue. Known positive controls were used in each hybridization reaction. These consisted of two cases for which positive hybridization was reproducible in pilot studies. Controls for endogenous alkaline phosphatase included treatment of the sample in the absence of the probe and use of chromogen alone.
Evaluation of mRNA ISH Results.
Staining was scored in carcinoma and stromal cells. Staining extent was scored as 0, 1, or 2, using a cutoff of 20%. Staining of 
20% of tumor/stromal cells was scored as focal (1)
, whereas staining of >20% of cells was interpreted as diffuse (2)
. Staining intensity was scored as absent (0), weak/moderate (1)
, or intense (2)
.
Statistical Analysis.
Staining values in tumor cells and stromal cells were evaluated statistically, applying the SPSS PC package (Version 10.1; SPSS, Chicago, IL). Probability of <0.05 was considered statistically significant. Analyses of the association between mRNA ISH results and biopsy site, patient group, tumor grade, and previously studied metastasis-associated molecules were executed using the two-sided 
2 test. Univariate survival analyses were executed using the Kaplan-Meier method and log-rank test. Both staining extent and intensity were analyzed. Multivariate analyses of survival were performed using the Cox regression model. The parameters included all previously studied markers that showed significance in univariate analysis of this cohort: mRNA expression of MMP-2, MMP-9, MT1-MMP, TIMP-2, Ets-1, and v integrin, as well as expression of the carbohydrate antigen sialyl Lewisx (33)
and the protein expression of the adhesion molecule 
-catenin (34)
. Clinical parameters (patient age, tumor type, grade of differentiation, and disease stage) were not included because their prognostic role was nullified by the choice of patients for this study.
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
.
|
. Tumor differentiation was as follows: 6 grade I tumors; 3 grade II tumors; and 27 grade III tumors. The fraction of poorly differentiated (grade III) tumors was comparable for both groups (11 of 16 and 16 of 20 tumors).
mRNA ISH
A positive signal using a poly d(T) probe was detected in all cases (Fig. 1A)
. Sections hybridized with the sense probe were uniformly negative, as were controls hybridized without probe (Fig. 1B)
. Expression of PEA3 mRNA was detected in carcinoma cells and peritumoral stromal cells in 56 of 61 lesions (92%) and 54 of 61 lesions (89%), respectively (Fig. 1, C-E)
. Labeling of tumor cells was intense in 5 specimens, whereas in 14 specimens, similar findings were seen in the stromal compartment. Labeling extent in tumor and stromal cells was categorized as diffuse in 45 specimens each.
|
. Intense stromal expression was detected only in the vicinity of grade 2–3 tumors (P = 0.04; Table 4
). Stromal expression was also more diffuse in tumors from older patients (P = 0.066; data not shown). PEA3 expression was comparable in primary and metastatic lesions and showed no association with FIGO stage (data not shown).
|
|
). Intense PEA3 mRNA expression in carcinoma cells showed an association with mRNA expression of ß1 integrin in the same compartment (P = 0.039; Table 6
; Fig. 1F
). It also showed association with intense mRNA expression of the ß1 integrin subunit (P = 0.012) and bFGF (P = 0.036; Fig. 1G
) and with diffuse labeling for EMMPRIN (P = 0.038) in stromal cells (Table 7
; Fig. 1H
).
|
|
|
). Intense expression in tumor cells similarly predicted poor overall survival (P = 0.049; Fig. 2C
). PEA3 mRNA expression in stromal cells emerged as an independent predictor of poor outcome in multivariate survival analysis, in which all molecules previously studied in this patient cohort were included (P = 0.015), together with TIMP-2 mRNA expression in stromal cells (P = 0.001) and the expression of MMP-9 (P = 0.046), sialyl Lewisx (P = 0.049), and -catenin (P = 0.013) in tumor cells.
|
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Early studies analyzing various benign and cancerous tissues have localized PEA3 mRNA and protein to carcinoma cells, whereas no expression was seen in benign tissue (17, 18 , 22 , 25) . PEA3 mRNA expression has been observed in >50% of breast, lung, and oral carcinomas (17 , 22 , 24) . In an additional report, 47% of tumors expressed PEA3 protein, but expression below 10% was scored as negative (18) . As opposed to early studies of Ets-1, stromal cell expression has not been documented. We found PEA3 expression in both tumor and stromal cells in the majority of the studied specimens, at both primary and metastatic sites. Furthermore, intense labeling was more frequently seen in the stromal compartment. The prominent expression in stromal cells is in agreement with the cellular localization of MMP-2, MMP-9, and TIMP-2 (28) , as well as that of angiogenic genes (30) , in this patient cohort. The central contribution of peritumoral fibroblasts to tumor progression has recently been reviewed (36) . Our data provide additional evidence supporting the dual (tumor and stromal cell) origin of metastasis-associated molecules in ovarian carcinomas.
The presence of PEA3 sites in MMP promoters and its role in activation of proteolytic enzyme expression are well established (6 , 7 , 14 , 23) . Furthermore, induction of collagenase transcription after stimulation by bFGF and integrins has been demonstrated in experimental models (37, 38, 39) . Coexpression of PEA3 and the angiogenic molecule interleukin 8 was shown in the HepG2 hepatoma cell line, as well as in three clinical specimens (40) . However, no data are available regarding the association between PEA3 expression and that of integrins, proteolytic enzymes, or other angiogenic genes in clinical specimens of human tumors. We found a significant association between mRNA expression of PEA3 and that of MMP-2, bFGF, and the ß1 integrin subunit in our patient cohort. Coexpression of EMMPRIN, a member of the immunoglobulin superfamily involved in MMP-1, MMP-2, and MMP-3 activation, with PEA3 was also seen. This finding provides the first evidence of a possible association between these molecules in vivo. Furthermore, these associations highlight the role of tumor-stromal collaboration in ovarian carcinoma.
Establishing the relevance and predictive role of molecular markers requires evaluation of well-defined patient cohorts. Our study evaluated two groups of patients diagnosed with advanced-stage ovarian carcinoma with a markedly different disease outcome, with a follow-up period of up to 20 years. Established prognostic factors, such as age, stage, grade, and tumor type, were all controlled by patient selection criteria in the design of the study. This selection was meant to facilitate the study of potential prognostic markers. Results recently published using this patient cohort identified Ets-1 (4)
; MMP-2, MMP-9, MT1-MMP, and TIMP-2 (28)
; the v integrin subunit (29)
; sialyl Lewisx antigen (33)
; the adhesion molecule -catenin (34)
; and topoisomerase II (41)
as prognostic markers in advanced-stage ovarian carcinoma.
The present study evaluated the role of PEA3 as a prognosticator in this cohort. Prognostic studies of PEA3 in carcinomas are limited to two studies to date. Hida et al. (24) reported more frequent expression of PEA3 in oral carcinomas (n = 27) showing invasive growth or lymph node metastasis. Evaluating 89 patients diagnosed with breast carcinoma, Kinoshita et al. (18) found protein expression of PEA3 in tumor cells to be a predictor of poor overall survival at 6 years. In the present study, intense expression of PEA3 mRNA in both tumor and stromal cells correlated with poor overall survival. Stromal cell labeling also predicted poor disease-free survival. Furthermore, PEA3 expression in stromal cells retained its prognostic power in multivariate analysis, providing the first evidence of an independent predictive role for this molecule in human malignancy. Interestingly, two of the most powerful markers in Cox analysis (TIMP-2 and PEA3) were localized to stromal cells. This finding provides decisive evidence in attributing a central biological role to stromal cells in ovarian carcinoma. Parenthetically, advanced patient age and high histological grade are two of the established clinical prognostic markers in ovarian carcinoma (42) . Although the predictive power of these parameters was nullified by the choice of patients for the long-term and short-term survivor groups, they did show an association with PEA3 expression.
Two tumors from short-term survivors showed no PEA3-positive cells in the tumor stroma, whereas intense expression was detected in stromal cells in two lesions from long-term survivors. Closer evaluation of these four cases revealed no unexpected data regarding clinical parameters. All four patients were diagnosed with FIGO stage III disease, had grade 3 tumors, and were about 60 years old. Similarly, the two short-term survivors with PEA3-negative tumors had an overall survival of 11 and 13 months, whereas the two long-term survivors with PEA3-positive tumors survived for 69 and 120 months, well in agreement with other patients’ data. MMP-2, MMP-9, MT1-MMP, and TIMP-2 mRNA expression was variable. However, of interest is the fact that these two PEA3-negative cases were also Ets-1 negative in our earlier report, whereas the two PEA3-positive tumors from long-term survivors were also Ets-1 positive. This suggests induction of MMP synthesis by other transcription factors in short-term survivors with Ets-1- and PEA3-negative tumors and the presence of other, possibly growth-suppressive targets for PEA3 and Ets-1 regulation in few tumors from long-term survivors.
Transcriptional control of metastasis-associated genes is one of the critical parameters affecting tumor progression in malignant diseases. Our recent studies of Ets-1 and the present study of PEA3 indicate a possible regulatory pattern of these genes. In view of the central role of MMP-2, bFGF, EMMPRIN, and integrins in tumor invasion and angiogenesis, their coexpression and colocalization with PEA3 offers a possible explanation about the pathways leading to PEA3-mediated activation of metastasis-associated gene expression.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
1 This study was presented at the 9th International Congress of the Metastasis Research Society, Chicago, Illinois, September 20–22, 2002. 
2 To whom requests for reprints should be addressed, at Department of Pathology, The Norwegian Radium Hospital, Montebello, N-0310 Oslo, Norway. Phone: 47-22934871; Fax: 47-22508554; E-mail: bend{at}ulrik.uio.no
3 The abbreviations used are: MMP, matrix metalloproteinase; ISH, in situ hybridization; bFGF, basic fibroblast growth factor; FIGO, International Federation of Gynecologists and Obstetricians; TIMP, tissue inhibitor of metalloproteinase.
4 B. Davidson, I. Goldberg, W. H. Gotlieb, J. Kopolovic, G. Ben-Baruch, and R. Reich, manuscript in preparation.
Received 6/25/02; revised 11/ 4/02; accepted 11/ 6/02.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
v integrin is a novel marker of poor prognosis in advanced-stage ovarian carcinoma. Clin. Cancer Res., 7: 4073-4079, 2001.This article has been cited by other articles:
|
|
 |
|
  J. Eo, K. Han, K. M Murphy, H. Song, and H. J. Lim Etv5, an ETS transcription factor, is expressed in granulosa and cumulus cells and serves as a transcriptional regulator of the cyclooxygenase-2 J. Endocrinol., August 1, 2008; 198(2): 281 - 290. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. D. Cowden Dahl, R. Zeineldin, and L. G. Hudson PEA3 Is Necessary for Optimal Epidermal Growth Factor Receptor-Stimulated Matrix Metalloproteinase Expression and Invasion of Ovarian Tumor Cells Mol. Cancer Res., May 1, 2007; 5(5): 413 - 421. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Kawamura-Saito, Y. Yamazaki, K. Kaneko, N. Kawaguchi, H. Kanda, H. Mukai, T. Gotoh, T. Motoi, M. Fukayama, H. Aburatani, et al. Fusion between CIC and DUX4 up-regulates PEA3 family genes in Ewing-like sarcomas with t(4;19)(q35;q13) translocation Hum. Mol. Genet., July 1, 2006; 15(13): 2125 - 2137. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. B. Koo, H. Song, I. Moon, K. Han, C. Chen, K. Murphy, and H. Lim Differential expression of the PEA3 subfamily of ETS transcription factors in the mouse ovary and peri-implantation uterus Reproduction, May 1, 2005; 129(5): 651 - 657. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Z.-Z. Su, D. Sarkar, L. Emdad, G. J. Duigou, C. S. H. Young, J. Ware, A. Randolph, K. Valerie, and P. B. Fisher Targeting gene expression selectively in cancer cells by using the progression-elevated gene-3 promoter PNAS, January 25, 2005; 102(4): 1059 - 1064. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Chotteau-Lelievre, F. Revillion, V. Lhotellier, L. Hornez, X. Desbiens, V. Cabaret, Y. de Launoit, and J.-P. Peyrat Prognostic Value of ERM Gene Expression in Human Primary Breast Cancers Clin. Cancer Res., November 1, 2004; 10(21): 7297 - 7303. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Liu, G. L. Borchert, and J. M. Phang Polyoma Enhancer Activator 3, an Ets Transcription Factor, Mediates the Induction of Cyclooxygenase-2 by Nitric Oxide in Colorectal Cancer Cells J. Biol. Chem., April 30, 2004; 279(18): 18694 - 18700. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
