Ovarian clear cell adenocarcinoma (OCCA) has been recognized as a distinct histological type of cancer in the World Health Organization classification of ovarian tumors.1 The incidence of OCCA among epithelial ovarian cancers is estimated to be less than 5% to 10%.2 However, OCCA occurs more frequently in Japan and Taiwan.3,4 It is known that OCCA represents a remarkably distinct clinical type compared with other epithelial ovarian cancers and shows a poor prognosis.2,5 Ovarian clear cell adenocarcinoma is also clinically characterized by de novo resistance to platinum-based chemotherapy.5,6
We have reported that complete surgical staging and paclitaxel plus carboplatin combinational chemotherapy seemed to improve the survival of patients with stage I as well as advanced OCCA.4,7 However, other reports state there is no survival improvement in both early and advanced OCCA using paclitaxel- and cisplatin-based regimens.
Prognoses for OCCA patients are considered not only by chemotherapeutic factors but also by tumor biologic factors. Studies in bladder, prostate, breast, lung, and ovary carcinomas also suggest that disturbance in epithelial cadherin (E-cadherin) function was not only important in carcinogenesis but may also be a critical event in tumor progression.8-10 E-cadherin, an adhesion protein, is considered a novel, promising prognostic factor in patients with ovarian cancer.10 Less is known, however, on the prognostic and predictive role of E-cadherin in patients with advanced OCCA who have received chemotherapy.
E-cadherin belongs to the cadherin family of calcium-dependent adhesion molecules and is mapped to chromosome 16q22.1.11 These molecules are transmembrane glycoproteins localized at the adherens junction of epithelial cells and mediate homotypic cell-to-cell adhesions.12 E-cadherin associates with the actin cytoskeleton through a group of membrane-associated proteins, α-catenin, β-catenin, and γ-catenin, which are essential for the maintenance of a stable E-cadherin-mediated cell adhesion and are involved in regulating cell motility.13 Loss of E-cadherin expression has been regarded as a central event in tumor metastasis, as loss of adhesion between tumor cells facilitates their ability to invade locally and to spread to distant organs.14
TP53 is a tumor suppressor gene located on the short arm of chromosome 17. Mutation or deletion of TP53 is believed to result in uncontrolled cell proliferation.15 Little is known about what roles E-cadherin and TP53 overexpression play in influencing the prognosis of OCCA. Lack of report on the prognostic role of E-cadherin expression in patients with advanced OCCA and the recent discovery of its potential role in determining chemosensitivity prompted us to investigate the predictive and the prognostic role of E-cadherin expression in patients with advanced OCCA who had received either paclitaxel-platinum or conventional platinum-based chemotherapy.16 In the present study, we sought to determine whether E-cadherin or TP53 immunoreactivity could predict survival in different types of regimens and also act as a prognostic indicator in treating advanced pure OCCA. This study may provide novel and useful information for planning treatment strategies and for improving the results of chemotherapy with OCCA.
MATERIALS AND METHODS
Between 1994 and 2006, a total of 71 patients with stages IIC, III, and IV pure cell-type OCCA were treated at 6 institutions in a project approved by the institutional review boards of each institution. Clinical data were collected from medical records and clinic visits. The pathology diagnosis of OCCA was reviewed by a pathologist (M.C.L.). After exclusion of 3 patients who refused adjuvant chemotherapy, patients with missing paraffin blocks, and 3 patients with incorrect histological types recorded, 62 patients (stage IIC, n = 5; IIIA, n = 1; IIIB, n = 4; IIIC, n = 42; IV, n = 10) with pure cell-type OCCA were included in this study. The agreement of 3 cases with incorrect histological types was achieved and reclassified as serous type after reviewing by 2 pathologists (M.C.L. and S.H.H.). None of these patients had undergone neoadjuvant chemotherapy before surgery. One patient with stage IIIC who received only a single cycle of chemotherapy was excluded in the analysis of chemoresponse and survival. Finally, 61 patients with advanced OCCA (stage IIC or greater) were recruited in this study.
Formalin-fixed, paraffin-embedded tissue samples were obtained from the departments of pathology at 6 institutions. All tissue specimens underwent microscopic confirmation of diagnosis and histological type by 2 independent gynecologic pathologists without knowledge of the clinical information. Ovarian clear cell adenocarcinoma was defined as typically clear cells or hobnail cells present in papillary, solid, or tubulocystic patterns. The staging of tumors was assigned according to the system of the International Federation of Gynecology and Obstetrics (FIGO).
All patients underwent comprehensive staging laparotomy and initial cytoreductive surgery followed by both paclitaxel and platinum chemotherapy defined as paclitaxel-based chemotherapy (n = 46) or conventional platinum chemotherapy defined as non-paclitaxel-based chemotherapy (n = 15). For the non-paclitaxel-based chemotherapy group, the patients received 75 to 100 mg/m2 of cisplatin and 750 to 1000 mg/m2 cyclophosphamide for the paclitaxel-based chemotherapy group; 41 of 46 patients received 175 mg/m2 of paclitaxel and carboplatin (area under the curve = 5), and the other 5 patients received 135 mg/m2 of paclitaxel and 75 mg/m2 of cisplatin. The median number of 6 cycles was given in both paclitaxel-based chemotherapy and non-paclitaxel-based chemotherapy groups (range, 2-12 cycles). Clinical response was assessed in patients with clinically measurable disease, according to World Health Organization criteria,17 or assessed in patients with nonmeasurable disease according to normal physical examinations, computed tomography of the abdomen or pelvis, chest x-ray, and CA-125 of less than 35 U/mL. Clinical progression was defined as new metastases, an increase in measurable disease by 50%, or an increase in CA-125 by 100%.
The clinical parameters studied were age, stage, CA-125, residual tumors after cytoreductive surgery (optimal vs suboptimal debulking), and type of regimens of chemotherapy.
For the immunohistochemistry, E-cadherin, clone NCH-38 and p53 protein, clone DO-7, and monoclononal antibodies were used from DAKO (Glostrup, Denmark). Paraffin sections, 4 μm thick, from a tumor block representative of the entire tumor were mounted on SuperFrost slides (Menzel-Gläser, Braunschweig, Germany) and dried overnight at 58°C. After deparaffination and rehydration, slides were immersed in Target Retrieval Solution for 10 minutes in 0.001M sodium citrate buffer (pH 6.0), and heat-induced (autoclave 121°C) epitope retrieval used. Tissue sections were incubated with primary antibody (monoclonal mouse anti-human E-cadherin, 100× dilute; or monoclonal mouse anti-human p53 protein, 100× dilute) for 30 minutes. Detection of primary antibody was performed using the appropriate biotin-streptavidin peroxidase detection system (I-VIEW DAB detection kit; Ventana, Mannheim, Germany). The staining procedure followed that of the detection system selected (NEXES IHC). The sections were counterstained with hematoxylin and mounted for microscopic examination.
Normal skin epithelium was used as a positive control with strong and homogenous expression of E-cadherin at the cell membrane throughout the epithelium. Normal skin epithelium without the primary antibody was used as a negative control. In all positive sections, p53 staining was localized to the cell nucleus. Negative control sections were those without exposed to the primary antibody. Immunohistochemical staining was assessed without prior knowledge of survival or other clinical data. Relationships between these results and clinicopathologic variables were also analyzed.
Evaluation and Quantification of Immunostaining
The E-cadherin immunoexpression of the tumors was scored semiquantitatively according to the percentage of positive tumor cells in membranous staining on a 4-point scale of 0 to 4 (0 = complete absence of expression, 1 = ≤10%, 2 = >10 and ≤33%, 3 = >33% and ≤66%, 4 = >66%). Negative E-cadherin immunoexpression was defined as 10% positive tumor cells or less, and positive E-cadherin immunoexpression was defined more than 10% positive tumor cells. The use of arbitrary 10% E-cadherin expression as a cutoff in the analysis was based on its reach in most significant difference in survival when compared with other percentage as a cutoff value. The p53 immunostaining was also semiquantified according to the following grading system: mild (from few positive tumor cells to <33% positive tumor cells), moderate (from ≥33% to <66% positive tumor cells), and strong (≥66% positive tumor cells). Negative p53 immunostaining was defined as less than 33%, and positive p53 immunostaining was defined as 33% or greater. The positive staining and negative staining of E-cadherin and p53 are shown in Figures 1A and B and 2A and B.
Statistical analysis was carried out using SPSS software version 8.0 (SPSS, Chicago, Ill). The association between negative versus positive E-cadherin expression and clinicopathologic parameters was evaluated using the χ2 test. The classification of E-cadherin immunoexpression pattern into negative vs positive expression associated significantly with overall survival (OS); therefore, this subdivision was used for further analysis regarding overall and progression-free survival (PFS). The patient population was compared using Fisher exact test and the Wilcoxon rank sum test. Survival time was considered the primary end point and was defined as from the date of diagnosis to the date of death or last contact. Survival information was available for all patients. Survival curves were generated using the Kaplan-Meier method, and differences in survival curves were calculated using the log-rank test. Cox univariate and multivariate regression analysis was used to evaluate prognostic factors for survival.
The characteristics of the 61 OCCA patients are summarized in Table 1. The median age of the 61 OCCA patients was 51 years (range, 32-72 years). Patients with OCCA had FIGO stages IIC (n = 5), IIIA (n = 1), IIIB (n = 4), IIIC (n = 41), and IV (n = 10). All patients were also followed up until last contact or death; the median follow-up was 22.7 months. For surviving patients, the median follow-up was 32.6 months (range, 2.1-83.0 months), and no one was lost to follow-up. The expected 5-year OS rate of these 61 OCCA patients was 30%.
p53 And E-Cadherin Immunoexpression in OCCA
The immunohistochemical staining of p53 and E-cadherin failed in 3 of the 61 patients because of poor quality of formalin-fixed, paraffin-embedded tissue samples. p53 Immunoexpression was observed in 44.8% (26/58) of OCCAs; in contrast, E-cadherin immunoexpression was observed in 75.9% (44/58) of OCCAs.
OCCA Patients Receiving Paclitaxel-Based Chemotherapy Had Better Survival Rates
The median survival for women with OCCA treated with paclitaxel-based chemotherapy was significantly longer than for those treated with non-paclitaxel-based chemotherapy (22.3 vs 9.0 months, P < 0.01), and as shown in Figure 2, the expected 5-year OS rate for OCCA women treated with paclitaxel-based chemotherapy was also significantly better than for those treated with non-paclitaxel-based chemotherapy (40% vs 0%, P = 0.001).
E-Cadherin, Not p53, Correlates With the Survival of OCCA Patients
For OCCA women stratified by E-cadherin immunoexpression, the expected 5-year OS rate was significantly better in patients with E-cadherin immunoexpression of greater than 10% than in those with E-cadherin immunoexpression of 10% or less (35% vs 0%, P = 0.02) (Fig. 3A). The percentage of patients with E-cadherin of 10% or less and greater than 10% in relation to tumor size, stage, optimal debulking or not, CA-125 of less than 500 U/mL, and chemoresponse, and receiving paclitaxel- or non-paclitaxel-based chemotherapy was not significantly different (all P > 0.05). When these 61 patients were stratified by p53 immunoexpression, the expected 5-year OS was not significantly different in patients with p53 immunoexpression of less than 33% than in those with p53 immunoexpression of 33% or greater (31% vs 28%, P = 0.39).
Fourteen patients with E-cadherin immunoexpression of 10% or less were stratified by paclitaxel-based versus non-paclitaxel-based chemotherapy; the expected 5-year OS of those receiving paclitaxel-based chemotherapy had not significantly benefited from those receiving non-paclitaxel-based chemotherapy (0% vs 0%, P = 0.11) (Fig. 3A). However, when evaluating 44 patients with E-cadherin immunoexpression of greater than 10%, the expected 5-year OS rate of those receiving paclitaxel-based chemotherapy was significantly better than those receiving non-paclitaxel-based chemotherapy (43% vs 0%, P = 0.01) (Fig. 3B).
E-Cadherin and Paclitaxel-Based Chemotherapy Are 2 Independent Prognostic Factors for OCCA Patients
Paclitaxel-based chemotherapy was the only favorable prognostic factor in the PFS of OCCA patients by univariate analysis (P < 0.001) (Table 2). However, E-cadherin immunoexpression of greater than 10% (P = 0.028), receiving paclitaxel-platinum-based chemotherapy (P < 0.001), and optimal cytoreductive surgery (P = 0.036) were 3 favorable prognostic factors in the OS of OCCA patients by univariate analysis (Table 3).
Paclitaxel-based chemotherapy (P = 0.015) was also the only favorable prognostic factor in PFS of OCCA patients by multivariate analysis (Table 2). However, E-cadherin immunoexpression of greater than 10% (P = 0.04) and paclitaxel-based chemotherapy (P = 0.01) were 2 independent favorable prognostic factors in the OS of OCCA patients by multivariate analysis (Table 3).
Paclitaxel-based chemotherapy could improve not only PFS but also OS in advanced OCCA patients. Our results are consistent and extended from our previous study,4 which showed that paclitaxel-based chemotherapy was the only prognostic factor for women with advanced OCCA in univariate analysis. In this study, we further demonstrated that paclitaxel-based chemotherapy could improve survival among patients with positive E-cadherin immunoreactivity in stage IIC-IV OCCA. To our knowledge, this is the first article to address using E-cadherin immunoreactivity may provide novel and useful information for planning treatment strategies and for improving the outcome of chemotherapy with OCCA.
Ovarian clear cell adenocarcinoma constitutes 5% to 10% of surface epithelial ovarian cancers, and about 30% to 40% of patients with OCCA have advanced-stage disease at diagnosis.18 Patients with advanced-stage OCCA represent only 3% to 4% of patients with stages III and IV ovarian malignancies with poorer prognoses than those with other pathological types of epithelial ovarian carcinoma.2 Standard treatment of OCCA is surgical resection of tumor mass, with complete reduction of the tumor, not being possible in a minor fraction of patients because of their advanced-stage disease. Most OCCA patients receive subsequent adjuvant chemotherapy, with a combination of taxane and platinum (eg, carboplatin, cisplatin) compounds representing the most frequently applied regimens. Taxanes stabilize microtubules, resulting in a G2-M cell cycle arrest and induction of apoptosis. Platinum drugs induce DNA cross-links, which are recognized by the DNA repair factors, and also result in apoptosis. Although the debulking status of the primary surgical resection is a potent prognostic factor in ovarian cancer, the role of debulking surgery in advanced OCCA has not been completely established4 and has not been proven in this study. So far, there are no other validated prognostic variables in advanced OCCA patients that allow patients' stratification for adjuvant treatment or that are associated with a predictive effect for the specific chemotherapeutic regimens.
Stage I OCCA tumors carried a relatively good prognosis,7 and p53 immunoexpression was observed in only 11.1% (1/9) (data not shown), whereas advanced OCCA tumors had a poorer prognosis, and p53 immunoexpression was observed with a higher incidence of 44.8% in this study. Based on these results, the question is raised if p53 mutations or overexpression drives metastatic behavior in advanced OCCA and contributes to their chemoresistance. However, our results did not support that loss of p53 expression was correlated with poor survival in patients with advanced OCCA.
Downexpression of E-cadherin expression relates to adverse outcomes in patients with ovarian carcinoma.10 Previous studies reported that a decreased immunoexpression pattern of E-cadherin was associated with clinical stage, lymph node metastasis, and differentiation.19 Negative E-cadherin and β-catenin have been shown to be poor prognostic factors in ovarian cancer.10 Expression of epithelial-mesenchymal transition-associated protein, such as Snail, Slug, SIP1, and Twist, has been reported to show an inverse correlation with that of E-cadherin in various malignancies.20 An unfavorable prognosis in E-cadherin negative ovarian cancer probably depends on upregulation of epithelial-mesenchymal transition-associated protein.21 There are some recent advances in molecular etiology in terms of pathogenesis in OCCA that demonstrated that OCCAs have a high frequency of activating PIK3CA mutations,22 and a more favorable outcome was observed for cystic OCCA compared with adenofibromatous OCCA in all stages.23
The significance of the E-cadherin in chemoresponse and prognosis remains unclear in advanced OCCA. The present study addressed the issue of whether downexpression of E-cadherin represents a clinically relevant marker for patients with advanced OCCA, in particular with respect to the prognostic and predictive value of E-cadherin for the survival of patients with different adjuvant chemotherapy protocols. Downexpression of E-cadherin may differentially affect taxane and non-taxane-based drugs in vitro.15 E-cadherin-mediated adhesion creates barriers to prevent the penetration of chemotherapy agents, and disruption of E-cadherin sensitized multicellular spheroids in vitro to treatment with paclitaxel but not cisplatin.15 In our study, for women with pure OCCA with negative E-cadherin immunoexpression (≤10%), the expected 3-year OS rate of patients receiving paclitaxel-platinum-based chemotherapy was significantly better than those receiving platinum-based chemotherapy (51% vs 0%, P = 0.04). It seems that paclitaxel itself has some antiadhesive activity that could overcome the E-cadherin-mediated adhesive barrier to penetration of chemotherapy agents. Unexpectedly, paclitaxel-platinum chemotherapy can improve survival among patients with positive E-cadherin in advanced OCCA. For women with pure OCCA with positive E-cadherin immunoexpression (>10%), who were composed of 76% of advanced OCCA patients, the expected 5-year OS rate of patients receiving paclitaxel-platinum-based chemotherapy was significantly better than those receiving platinum-based chemotherapy (43% vs 0%, P = 0.01). It seems that paclitaxel has a remarkable antitumor effect on patients with positive E-cadherin in advanced OCCA.
The effects of E-cadherin expression can be applied as a clinically relevant prognostic and predictive marker to identify subsets of patients who could, to some degree, benefit from different regimens of chemotherapy as shown in this study. We disclose that downimmunoexpression of E-cadherin carried a worse prognosis, whereas receiving paclitaxel/cisplatinum therapy was associated with improved OS in patients with positive E-cadherin in advanced OCCA. This is comparable with a previous study that showed that the expression of E-cadherin could carry a better prognosis in primary epithelial ovarian cancer.18 Our results suggest that paclitaxel-platinum chemotherapy could improve survival in advanced OCCA patients, especially for those with positive E-cadherin. These results underline the necessity for investigating candidate predictive markers in advanced OCCA patients with different adjuvant treatment regimens.
1. Serov SF, Scully RE, Robin IH. International histological classification of tumors. Histological Typing of Ovarian Tumours, No. 9
. Geneva, Switzerland: World Health Organization; 1973:37-42.
2. O'Brien ME, Schofield JB, Tan S, et al. Clear cell epithelial ovarian cancer (mesonephroid): bad prognosis only in early stages. Gynecol Oncol
3. Ueki M. Report of gynecologic cancer in Japan. Acta Obstet Gynecol Jpn
4. Ho CM, Huang YJ, Chen TC, et al. Pure-type clear cell carcinoma
of the ovary as a distinct histological type and improved survival in patients treated with paclitaxel
in pure-type advanced disease. Gynecol Oncol
5. Sugiyama T, Kamura T, Kigawa J, et al. Clinical characteristics of clear cell carcinoma
of the ovary: a distinct histological type with poor prognosis and resistance to platinum-based chemotherapy
6. Montag AG, Jenison EL, Griffiths CT, et al. Ovarian clear cell carcinoma
. A clinicopathologic analysis of 44 cases. Int J Gynecol Pathol
7. Ho CM, Chien TY, Shih BY, et al. Evaluation of complete surgical staging with pelvic and para-aortic lymphadenectomy and paclitaxel
plus carboplatin chemotherapy
for improvement of survival in stage I ovarian clear cell carcinoma
. Gynecol Oncol
8. Bringuier PP, Umbas R, Schaafsma HE, et al. Decreased E-cadherin
immunoreactivity correlates with poor survival in patients with bladder tumors. Cancer Res
9. Bremnes RM, Veve R, Gabrielson E, et al. High-throughput tissue microarray analysis used to evaluate biology and prognostic significance of the E-cadherin
pathway in non-small-cell lung cancer. J Clin Oncol
10. Faleiro-Rodrigues C, Macedo-Pinto I, Pereira D, et al. Prognostic value of E-cadherin
immunoexpression in patients with primary ovarian carcinomas. Ann Oncol
11. Hajra KM, Fearon ER. Cadherin and catenin alterations in human cancers. Genes Chromosomes Cancer
12. Takeichi M. Cadherins. A molecular family important in selective cell-cell adhesion. Annu Rev Biochem
13. Hinck L, Näthke IS, Papkoff J, et al. Dynamics of cadherin/catenin complex formation: novel protein interactions and pathways of complex assembly. J Cell Biol
14. Frixen UH, Behrens J, Sachs M, et al. E-cadherin
mediated cell-cell adhesion prevents invasiveness of human carcinoma cells. J Cell Biol
15. Levine AJ, Momand J, Finlay CA. The p53 tumor suppressor gene. Nature
16. Nakamra T, Kato Y, Fujii H, et al. E-cadherin
dependent intercellular adhesion enhances chemoresistance. Int J Mol Med
17. WHO Handbook for Reporting Results of Cancer Treatment
. Geneva, Switzerland: World Health Organization; 1979.
18. Omura GA, Brady MF, Homesley HD, et al. Long-term follow-up prognostic factor analysis in advanced ovarian carcinoma
: the Gynecologic Oncology Group experience. J Clin Oncol
19. Chekhun VF, Lukyanova NY, Urchenko OV, et al. The role of expression of the components of proteome in the formation of molecular profile of human ovarian carcinoma
A2780 cells sensitive and resistant to cisplatin. Exp Oncol
20. Lee TK, Poon RT, Yuen AP, et al. Twist overexpression correlates with hepatocellular carcinoma metastasis through induction of epithelial-mesenchymal transition. Clin Cancer Res
21. Yoshida J, Horiuchi A, Kikuchi N, et al. Changes in the expression of E-cadherin
repressors, Snail, Slug, SIP1, and Twist, in the development and progression of ovarian carcinoma
: the important role of Snail in ovarian tumorigenesis and progression. Med Mol Morphol
22. Kuo KT, Mao TL, Jones S, et al. Frequent activating mutations of PIK3CA in ovarian clear cell carcinoma
. Am J Pathol
23. Veras E, Mao TL, Ayhan A, et al. Cystic and adenofibromatous clear cell carcinomas of the ovary: distinctive tumors that differ in their pathogenesis and behavior: a clinicopathologic analysis of 122 cases. Am J Surg Pathol
Keywords:Copyright © 2010 by IGCS and ESGO
Ovarian carcinoma; Clear cell carcinoma; Paclitaxel; Chemotherapy; E-cadherin