Human Pathology (2010) 41, 1205–1209
www.elsevier.com/locate/humpath
Original contribution
Implications of enhancer of zeste homologue 2 expression in pancreatic ductal adenocarcinoma☆ Adam D. Toll MD a , Abhijit Dasgupta PhD b , Magdalena Potoczek BS a , Charles J. Yeo MD c , Celina G. Kleer MD d , Jonathan R. Brody PhD a,c , Agnieszka K. Witkiewicz MD a,⁎ a
Department of Pathology, Thomas Jefferson University, Jefferson Pancreas, Biliary and Related Cancer Center, Philadelphia, PA 19107, USA b Department of Pharmacology and Experimental Therapeutics, Thomas Jefferson University, Jefferson Pancreas, Biliary and Related Cancer Center, Philadelphia, PA 19107, USA c Department of Surgery, Thomas Jefferson University, Jefferson Pancreas, Biliary and Related Cancer Center, Philadelphia, PA 19107, USA d Department of Pathology, University of Michigan, Ann Arbor, MI 48109, USA Received 17 November 2009; revised 4 March 2010; accepted 5 March 2010
Keywords: EZH2; Pancreatic ductal adenocarcinoma
Summary Pancreatic ductal adenocarcinoma is the fourth leading cause of cancer deaths in the United States. Single-agent gemcitabine remains the standard treatment of advanced pancreatic adenocarcinoma. A recently discovered histone methyltransferase termed enhancer of zeste homologue 2 (EZH2) was found to be overexpressed in a variety of carcinomas including pancreatic adenocarcinoma. Silencing of E-cadherin was proposed as a mechanism by which enhancer of zeste homologue 2 mediates tumor aggressiveness, and enhancer of zeste homologue 2 depletion has been found to sensitize pancreatic cancer cells to gemcitabine. In this study, we correlated enhancer of zeste homologue 2 with E-cadherin expression in pancreatic adenocarcinoma and evaluated response to gemcitabine in relation to enhancer of zeste homologue 2 expression in tumor cells. Fifty-four pancreatic adenocarcinomas, 13 intraductal papillary mucinous neoplasms, and 6 chronic pancreatitis cases were stained with antibodies against enhancer of zeste homologue 2 and E-cadherin. Enhancer of zeste homologue 2 staining was scored from 1 to 4+ and classified as either low (1-2+ in b25% of tumor nuclei) or high (3-4+ in N25% of tumor nuclei). E-cadherin expression was scored on membrane positivity as follows: 0 (0%-10%), 1 (10%-25%), 2 (25%-75%), and 3 (N75%). High enhancer of zeste homologue 2 expression in pancreatic adenocarcinoma was significantly associated with decreased Ecadherin expression and more aggressive disease. There was significantly longer survival in gemcitabine-treated patients with low versus high enhancer of zeste homologue 2 expression. High enhancer of zeste homologue 2 expression was detected in intraductal papillary mucinous neoplasms with moderate to severe dysplasia, but not in chronic pancreatitis. Our study suggests that E-cadherin down-regulation may lead to enhancer of zeste homologue 2–mediated invasion and metastasis. © 2010 Elsevier Inc. All rights reserved.
☆ The authors have no financial disclosures. ⁎ Corresponding author. Department of Pathology, Thomas Jefferson University Hospital, Philadelphia, PA, USA. E-mail address:
[email protected] (A. K. Witkiewicz).
0046-8177/$ – see front matter © 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.humpath.2010.03.004
1206
A. D. Toll et al.
Pancreatic ductal adenocarcinoma (PDA) is the fourth leading cause of cancer death in the United States with more than 35,000 annual deaths [1]. Surgical resection is the only potentially curative treatment. However, only a minority of patients meet the criteria for surgery; and thus, more than 80% of patients have metastatic or unresectable disease at the time of diagnosis. Single-agent gemcitabine remains the standard treatment of advanced PDA, which has shown improvement in disease-related symptoms and a modest benefit in survival [2]. More recently, gemcitabine has been used in combination with targeted therapies. Based on results from recent clinical trials, gemcitabine in combination with epidermal growth factor receptor inhibitor is considered the standard of care for advanced PDA patients in North America [2]. Epigenetic silencing of tumor suppressor genes has become a recognized part of tumorigenesis and is a recognized hallmark of pancreatic cancer [3,4]. Gene expression regulation through epigenetic silencing occurs through either DNA methylation or posttranslational modification of histone proteins [5]. One family of polycomb group proteins is capable of methylating both DNA and core histones. These proteins temporally and spatially restrict tissue stem cells while maintaining proliferative potential and establishing terminal differentiation patterns [6]. Polycomb group complexes contain a histone methyltransferase enhancer, zeste homologue 2 (EZH2), which has previously been shown to be overexpressed in a variety of carcinomas including PDA [5]. Furthermore, EZH2 depletion in pancreatic cancer cell lines rendered cells sensitive to gemcitabine by inducing apoptosis [5]. Additional work demonstrated that EZH2 mediates transcriptional silencing of the tumor suppressor gene Ecadherin, and inhibitors of EZH2 can attenuate malignant transformation in breast and prostate cancer cell lines [7]. In this study, we examined EZH2 and E-cadherin expression in PDA and correlated the expression of these candidate biomarkers with patient survival and response to gemcitabine treatment. In addition, we evaluated EZH2 expression in tissues from precursor lesions to PDA, intraductal papillary mucinous neoplasms (IPMNs), as well as chronic pancreatitis.
formalin-fixed, paraffin-embedded 4-μm sections. Representative sections were stained with both EZH2 (BD Bioscience, 1:25) and E-cadherin (1:25; BD Bioscience, San Jose, CA, USA). Staining procedures followed a previously published protocol [8]. Analysis of EZH2 expression was modeled after previous work [9,10] and scored blindly by an experienced pathologist (A. W.). Nuclear EZH2 expression was scored by using a validated system [11-13] as negative (score = 1, no staining), weak (score = 2, b 25% of nuclei staining, any intensity), moderate (score = 3, 25%-75% of nuclei staining, any intensity), and strong (score = 4, N75% of nuclei staining, any intensity). High EZH2 was defined as scores 3 and 4; low EZH2 was defined as scores 1 and 2. Ecadherin expression was evaluated according to the percentage of cells showing membrane positivity: score 0, 0% to 10% cells staining; score 1, 10% to less than 25%; score 2, 25% to 75%; and score 3, greater than 75% staining. Expression of E-cadherin was considered normal when scores were 3, reduced when equal to 2, and negative when scores were 1 or 0. The IPMN and chronic pancreatitis cases were stained only for EZH2 and scored in a manner similar to PDA. Association of EZH2 expression with E-cadherin expression, pathologic characteristics of the PDA, and demographic characteristics of the patients was assessed using Fisher exact test or Wilcoxon rank sum test depending on whether the variable was categorical or continuous. Overall survival of patients receiving only gemcitabine postoperatively was stratified by EZH2 expression and assessed using a KaplanMeier plot and log-rank test. Multivariate analysis of the effect of EZH2 on overall survival among patients receiving adjuvant gemcitabine treatment was assessed using Cox proportional hazards regression, adjusting for E-cadherin expression, tumor size, R stage, T stage, and N stage. After surgery, patients were followed for up to 48 months or until death (mean, 21.8 months). Adjuvant treatment modalities included the following: gemcitabine (n = 32), gemcitabine + carboplatin (n = 4), gemcitabine + capecitabine (n = 1), gemcitabine + capecitabine + radiation (n = 2), gemcitabine + radiation (n = 4), radiation (n = 2), capecitabine (n = 1), and no treatment (n = 8). The validity of the proportional hazards assumption was also assessed. P values b .05 were considered significant.
2. Methods
3. Results
Fifty-four surgically resected PDAs, 13 IPMNs (5 cases with mild dysplasia, 5 with moderate dysplasia, and 3 with severe dysplasia), and 6 chronic pancreatitis cases were obtained from the Department of Surgical Pathology of Thomas Jefferson University Hospital. Eight IPMNs were intestinal type, 1 was gastric, and 4 were pancreaticobiliary types. Immunohistochemical staining was performed on
3.1. Association between EZH2 and E-cadherin expression in PDA
1. Introduction
High EZH2 expression in PDA was associated with decreased E-cadherin expression in 70% of cases (compared with 35% for low EZH2 expression), node positivity (86% versus 48%), and larger tumor size (median of 4 versus 2.5
EZH2 in pancreatic cancer Table 1
1207
Descriptive statistics by EZH2 expression
Age (y) Sex Female Male T stage T1/T2 T3/T4 N stage N0 N1 Grade 1 2 3 Presence of microscopic residual disease R0 R1 Tumor size (mm) Extrapancreatic extension No Yes
Low (n = 23)
High (n = 31)
61
67
52% (12) 48% (11)
28% (8) 72% (21)
35% (8) 65% (15)
17% (5) 83% (24)
52% (12) 48% (11)
14% (4) 86% (25)
17% (4) 74% (17) 9% (2)
7% (2) 66% (19) 28% (8)
P value .617 .09
.201
.006
.175
1 87% (20) 13% (3) 2.5
84% (26) 16% (5) 4
35% (8) 65% (15)
17% (5) 83% (24)
.026 .201
cm) (Table 1, Fig. 1). All well-differentiated PDAs showed normal expression of E-cadherin. Moderate and poorly differentiated PDAs showed normal E-cadherin expression in 44% and 36% of cases, respectively. Other poor prognostic parameters including extrapancreatic extension and presence of microscopic residual disease (20.3 months of survival for R0 and 13.2 months for R1) were also associated with high EZH2 expression (Table 1). The difference in overall survival between high and low EZH2 expression in tumors among patients receiving only gemcitabine treatment (n = 32) was statistically significant by the log-rank test (Fig. 2).
3.2. EZH2 expression in IPMN and chronic pancreatitis Among the cases of IPMN and chronic pancreatitis, 85% of IPMN cases (11/13) and 33% of chronic pancreatitis cases (2/6) showed some degree of EZH2 expression. No cases of IPMN with mild dysplasia demonstrated EZH2 overexpression (0/5). Moderate dysplasia was associated with EZH2 overexpression in 40% (2/5) of cases, and severe dysplasia showed overexpression in 66% (2/3) of cases. Among 6 cases of chronic
Fig. 1 A-F, EZH2 and E-cadherin expression in well-differentiated PDA (A, hematoxylin-eosin, original magnification ×200). Strong membranous E-cadherin expression in PDA (B, immunostain, original magnification ×200). Focal nuclear expression of EZH2 in PDA (C, immunostain, original magnification ×200). EZH2 and E-cadherin expression in poorly differentiated PDA (D, hematoxylin-eosin, original magnification ×200). Strong membranous E-cadherin expression in a benign duct with reduced staining in PDA (E, immunostain, original magnification × 200). Diffuse nuclear expression of EZH2 in PDA with no staining in a benign duct (F, immunostain, original magnification × 200).
1208
Fig. 2 Kaplan-Meier survival curve showing low EZH2 expression in PDA is associated with a significant survival benefit relative to high EZH2 expression in patients treated with only gemcitabine (P = .001).
pancreatitis, 2 showed low expression of EZH2 expression and 4 were negative.
4. Discussion The results of our study demonstrate that EZH2 overexpression in PDA is associated with low E-cadherin expression, increased node positivity, and larger tumor size. The EZH2 expression was related to the degree of dysplasia in IPMN. High EZH2 expression was seen only in IPMN cases with at least moderate dysplasia. Chronic pancreatitis was essentially negative or showed low EZH2 expression. Together, these findings suggest that increased EZH2 expression levels may be an early, important event in pancreatic tumorigenesis. Previous work has shown an association between EZH2 expression and increased aggressiveness in a variety of carcinomas [8,10,14]. In breast cancer, EZH2 overexpression was associated with anchorage-independent growth and cellular invasion corresponding to higher stage and decreased survival [10]. Studies on high-stage prostate cancer used fluorescent in situ hybridization to demonstrate increased EZH2 copy number in tumor cells [8]. This was associated with higher rates of locally recurrent and metastatic hormone-refractory tumors. In melanocytic lesions, EZH2 expression increased incrementally from of benign nevi to malignant melanoma [14]. Similarly, in our study, we found that expression of EZH2 in IPMN increased with the degree of dysplasia (Fig. 3). Of note, EZH2 overexpression has not always been associated with worse clinical outcomes. Hinz et al [15] studied renal cell
A. D. Toll et al. carcinoma and found that high tumor EZH2 levels were associated with less aggressive disease. Our work coincides with the only previous work that studied EZH2 as it relates to PDA. Ougilkov et al [5] reported that high nuclear EZH2 expression was detected in pancreatic cancer cell lines and EZH2 expression levels correlated with higher tumor grade in human PDA samples. This study also reported that RNA interference silencing of EZH2 sensitized PDA to gemcitabine. We evaluated EZH2 expression in tumor samples from pancreatic adenocarcinoma patients treated with gemcitabine and found statistically significant differences in survival using the log-rank test (Fig. 2). The 2-year survival rate was 60% for patients with low EZH2 and 17.4% for patients with high EZH2, indicating a strong association between EZH2 levels and overall survival. Furthermore, we were unable to confirm a significant association between EZH2 expression and higher tumor grade; but we did demonstrate that high EZH2 expression correlated with higher disease stage. Recently, it has also been shown that EZH2 silences the tumor suppressor gene E-cadherin via methylation of histone H3-lysine 27 in prostate and breast cancer cell lines [7]. Reduced E-cadherin expression is associated with metastases and advanced tumor stage in several cancers [7,16]. In PDA, loss of E-cadherin protein is associated with poorly differentiated, noncohesive neoplasms [17]. Our study is the first to report that EZH2 overexpression is associated with decreased E-cadherin expression in PDA. The ability to silence E-cadherin could explain the seemingly contradictory association of EZH2 overexpression with aggressive disease, whereas its biologic role contributes to cell-cycle arrest. Future studies will reveal whether EZH2 is regulating Ecadherin through its ability to modify core histones. In summary, we showed that high EZH2 in PDA is associated with higher tumor stage and worse clinical outcome in patients treated with gemcitabine. As EZH2 functions to promote oncogenesis, the development of a small molecule inhibitor would be of great importance in the future management of PDA.
Fig. 3 A-B, Diffuse EZH2 expression in IPMN with severe dysplasia (A, hematoxylin-eosin, original magnification ×200; B, immunostain, original magnification ×200). Cases with mild dysplasia were negative.
EZH2 in pancreatic cancer
References [1] Jemal A, Siegel R, Ward E, et al. Cancer statistics, 2008. CA Cancer J Clin 2008;58:71-96. [2] Moore MJ, Goldstein D, Hamm J, et al. Erlotinib plus gemcitabine compared with gemcitabine alone in patients with advanced pancreatic cancer: a phase III trial of the National Cancer Institute of Canada Clinical Trials Group. J Clin Onco 2007;25:1960-6. [3] Baylin SB, Esteller M, Rountree MR, et al. Aberrant patterns of DNA methylation, chromatin formation and gene expression in cancer. Hum Mol Genet 2001;10:687-92. [4] Sato N, Fukushima N, Maitra A, et al. Discovery of novel targets for aberrant methylation in pancreatic carcinoma using high-throughput microarrays. Cancer Research 2003;63:3735-42. [5] Ougolkov AV, Bilim VN, Billadeau DD. Regulation of pancreatic tumor cell proliferation and chemoresistance by the histone methyltransferase enhancer of zeste homologue 2. Clin Cancer Res 2008;14: 6790-6. [6] Ezhkova E, Pasolli HA, Parker JS, et al. EZH2 orchestrates gene expression for the stepwise differentiation of tissue-specific stem cells. Cell 2009;136:1122-35. [7] Cao Q, Yu J, Dhanasekaran SM, et al. Repression of E-cadherin by the polycomb group protein EZH2 in cancer. Oncogene 2008;27:7274-84. [8] Saramäki OR, Tammela TLJ, Martikainen PM, et al. The gene for polycomb group protein enhancer of zeste homolog 2 (EZH2) is amplified in late-stage prostate cancer. Gene Chromosome Cancer 2006;45:639-45.
1209 [9] Collett K, Eide GE, Arnes J, et al. Expression of enhancer of zeste homologue 2 is significantly associated with increased tumor cell proliferation and is a marker of aggressive breast cancer. Clin Cancer Res 2006;12:1168-74. [10] Kleer CG, Cao Q, Varambally S, et al. EZH2 is a marker of aggressive breast cancer and promotes neoplastic transformation of breast epithelial cells. PNAS 2003;100:11606-11. [11] Raaphorst FM, Meijer CJ, Fieret E, et al. Poorly differentiated breast carcinoma is associated with increased expression of the human polycomb group EZH2 gene. Neoplasia 2003;5:481-8. [12] Bracken AP, Pasini D, Capra M, et al. EZH2 is downstream of the pRB-E2F pathway, essential for proliferation and amplified in cancer. EMBO J 2003;22:5323-35. [13] Varambally S, Dhanasekaran SM, Zhou M, et al. The polycomb group protein EZH2 is involved in progression of prostate cancer. Nature 2002;419:624-9. [14] McHugh JB, Fullen DR, Ma L, et al. Expression of polycomb group protein EZH2 in nevi and melanoma. J Cutan Pathol 2007;34: 597-600. [15] Hinz S, Weikert S, Magheli A, et al. Expression profile of the polycomb group protein enhancer of zeste homologue 2 and its prognostic relevance in renal cell carcinoma. J Urol 2009;182:2920-5. [16] Frixen UH, Behrens J, Sachs M, et al. E-cadherin-mediated cell-cell adhesion prevents invasiveness of human carcinoma cells. J Cell Biol 1991;113:173-85. [17] Winter JM, Ting AH, Vilardell F, et al. Absence of E-cadherin expression distinguishes noncohesive from cohesive pancreatic cancer. Clin Cancer Res 2008;14:412-8.
