J Neurooncol DOI 10.1007/s11060-014-1459-5
LABORATORY INVESTIGATION
The overexpression of Epithelial cell adhesion molecule (EpCAM) in glioma Xin Chen • Wei-Yuan Ma • Shang-Chen Xu • Yu Liang • Yi-Bing Fu Bo Pang • Tao Xin • Hai-Tao Fan • Rui Zhang • Jian-Gang Luo • Wen-Qing Kang • Min Wang • Qi Pang
•
Received: 6 August 2013 / Accepted: 23 April 2014 Ó Springer Science+Business Media New York 2014
Abstract Epithelial cell adhesion molecule (EpCAM) is overexpressed in various neoplasms as a tumor-associated antigen and absent in natural brain. However, little is known about EpCAM’s expression in gliomas. To investigate the expression of EpCAM in gliomas and understand the correlation of EpCAM expression with malignancy, proliferation, angiogenesis, and prognosis, we studied the expression of EpCAM in 98 glioma samples by immunohistochemistry and by western blotting (N = 12). Correlative analysis of EpCAM overexpression with microvessel density (MVD), Ki-67 expression, age, and gender were made. Survival data was analyzed with Kaplan–Meier method and Cox Proportional Hazard Model. Immunohistochemistry results showed EpCAM was widely expressed in glioma (90.8 %). The overexpression rate of WHO grade IV gliomas was significantly higher EpCAM
overexpression correlated significantly with Ki-67 expression and MVD. Western blot analysis also revealed a stepwise increase in EpCAM expression from WHO II to IV glioma. The overall survival of WHO III and IV glioma patients with EpCAM overexpression was obviously lower than that without EpCAM overexpression. EpCAM overexpression was an independent prognostic factor for overall survival in glioma patients. This study firstly shows that EpCAM overexpression correlates significantly with malignancy (WHO grades), proliferation (Ki67), angiogenesis (MVD), and prognosis in gliomas. EpCAM may participate in tumorgenesis of gliomas. Keywords Glioma Epithelial cell adhesion molecule (EpCAM) Overexpression Tumorigenesis
Introduction X. Chen S.-C. Xu T. Xin H.-T. Fan R. Zhang Q. Pang (&) Department of Neurosurgery, Shandong Provincial Hospital Affiliated to Shandong University, Jinan 250021, People’s Republic of China e-mail: [email protected] X. Chen W.-Y. Ma Y. Liang B. Pang R. Zhang J.-G. Luo W.-Q. Kang Shandong University School of Medicine, Jinan 250012, People’s Republic of China Y.-B. Fu Department of Gynaecology, Shandong Provincial Hospital Affiliated to Shandong University, Jinan 250021, People’s Republic of China M. Wang (&) Department of Neurology, Yantai Mountain Hospital, Yantai 264001, People’s Republic of China e-mail: [email protected]
Glioma, the most common neoplasm of central nervous system, accounts for more than 70 % of brain tumors [1]. According to the World Health Organization (WHO) classification, the most used glioma grading system [2], gliomas are subdivided to four grades (I–IV) based on the degree of malignancy. According to the WHO grades, I and II gliomas are generally defined to be benign, and gliomas of higher grades (III–IV), referred to malignant glioma, is more invasive and has worse outcomes [3, 4]. Although the WHO classification for glioma is accepted widely, but it is far from ideal [5]. Recent studies have identified new diagnostic and prognostic markers to supplement the WHO histological classification, such as the 1p19q co-deletion, O6-methylguanine DNA methyltransferase status, and mutations of isocitrate dehydrogenases 1 and 2 [6]. However, the understanding about
123
J Neurooncol
the pathogenesis of glioma is still largely unclear and the prognosis of patients with malignant glioma is still poor despite optimal treatments which combine surgery, chemotherapy, and irradiation [1, 3, 6, 7]. There thus is an urgent need to identify novel molecular markers for diagnosis and as potential therapeutic targets for glioma. Epithelial cell adhesion molecule (EpCAM) is a glycosylated, type I transmembrane protein, which is expressed in human cancers, progenitor and stem cells [8–10]. EpCAM was initially reported as a dominant surface antigen on human colon carcinoma [11], further comprehensive analysis of EpCAM expression has shown that EpCAM is expressed in many human cancers [12]. Because of the tumor-specific expression of EpCAM, it has been explored as a prognostic/diagnostic marker and anti-cancer target for immunotherapeutic strategies since 1970s [12–14]. However, it was not until the twentieth century that the roles of EpCAM in the intracellular signaling pathways for proliferation, migration, and mitogenic signal transduction were revealed [15, 16]. We sought to understand the role of EpCAM in tumorigenesis of glioma. The aim was to investigate the relationship of WHO classification of gliomas with some clinicopathologic factors, and prognosis. The results showed that EpCAM is widely expressed in gliomas and revealed a statistically significant correlation of EpCAM overexpression with the WHO classification, proliferation factor ki-67, microvessel density (MVD), and prognosis.
Table 1 Clinicopathologic characteristics of patients and Immunohistochemical expression of EpCAM in gliomas with different WHO classification Variables
III (25)
IV (49)
Total (98)
No expression(TIS 0)
4
3
2
9
Weak expression(TIS 1–4)
16
10
11
37
Moderate expression(TIS 6,8)
3
7
10
20
Intense expression(TIS 9,12)
1
5
26
32
Male
8
14
29
51
Female
16
11
20
47
P*
-0.161
0.113
NO overexpression
Overexpression
Gender
Extent of resection [90 %
18
14
20
52
\90 %
6
11
29
46
None
1
1
2
4
Radiotherapy only
18
7
6
31
Chemotherapy only
0
0
2
2
Radiotherapy and chemotherapy combination
5
17
39
61
Adjuvant treatment
Patients and samples
123
II (24)
ra
EpCAM overexpression
Materials and methods
We analyzed the archival records from the collection of glioma files of Provincial Hospital Affiliated to Shandong University, Jinan, P. R. China, and selected 98 cases from consecutive patients according to the criteria:1 Undergoing surgeries from 2003 to 2013; 2 Agreeing to contribute to this study and signing consents; 3 Preoperative Karnofsky performance score score [70; 4 None of the patients had received chemotherapy or radiotherapy prior to surgery. The Clinical data (including grade, age, gender and extent of resection) were obtained from patients’ medical records or attending physicians (Table 1). Contacts, adjacent treatment, and survival information were obtained from hospital information systems and the patients or patients’ relatives. Sections of all tumors were produced from paraffin-embedded tissue samples stored in Department of Pathology and reviewed by 2 pathologists to define the histological grade. 12 fresh glioma specimens were collected and stored by the Institute of Neurosurgery in Provincial Hospital Affiliated to Shandong University and
WHO grades (number)
Age(x)
38.2
47.0
51.8
47.3
0.174
0.086
Ki-67(x)
7.6
23.9
39.9
27.9
0.475
0.000*
MVD(x)
34.6
44.1
54.3
46.8
0.226
0.026*
WHO World Health Organization, MVD microvessel density, x mean value, EpCAM epithelial cell adhesion molecule, TIS total immunostaining score a
Spearman rank correlation coefficient (r) is used to analyze the correlation of EpCAM with clinicopathological parameters * P values indicating significance when P \ 0.05
diagnosed by Departments of Pathology. In this study, all the operations have been performed in accordance with the ethical standards and all the experiments comply with the current laws of China. Immunohistochemistry The immunohistochemical study was performed using the Envision PV-style two-step method (PV-9000 Polymer Detection System, Zhongshan Goldenbridge Biotechnology, China). A section of intestinal adenocarcinoma with known EpCAM positivity [12] was used as a positive control. For the negative control, primary antibodies were
J Neurooncol
replaced by the preparation solution for antibodies. Formalin-fixed, paraffin-embedded tissue sections of 4-lM thickness were baked at 60 °C for 30 min, deparaffinized in xylene, and rehydrated in graded concentrations of ethanol. Heat-induced antigen retrieval (10 mM citrate buffer [pH 6.0] at 98 °C for 20 min in a thermostatcontrolled waterbath) was carried out. Endogenous peroxidase activity was quenched by incubating in 0.3 % hydrogen peroxide at 37 °C for 30 min. Nonspecific staining was blocked by normal serum from the same species as that of secondary antibodies at 37 °C for 30 min. Primary antibodies (anti-EpCAM, ab71916, Abcam, Cambridge, UK, 1/100; anti-Ki67, ab66155, Abcam, Cambridge, UK, 1/200; anti-CD34, ab110643, Abcam, Cambridge, UK, 1/200) were applied at 4 °C overnight in a humidified chamber, then rewarming at 37 °C for 30 min, followed by Polymer Helper (Zhongshan Goldenbridge Biotechnology) incubation at 37 °C for 20 min and polyperoxidase-anti-mouse/rabbit IgG (Zhongshan Goldenbridge Biotechnology) incubation at 37 °C for 25 min. Diaminobenzidine was used as the enzyme substrate to observe the specific antibody localization, and hematoxylin was used as a nuclear counterstain. Sections were examined and scored for EpCAM, MVD and Ki-67 by two observers who were unaware of the histological diagnoses or clinical features. All the samples were stained at the same time. EpCAM expression for glioma staining was evaluated by calculating a total immunostaining score (TIS) as the product of a proportion score (PS) and an intensity score (IS). The PS describes the estimated fraction of positively stained cells (0, none; 1, \10 %; 2, 10–50 %; 3, 51–80 %; 4, [80 %). The IS represents the estimated staining intensity as compared to the controls (0 no staining; 1 weak; 2 moderate; 3 strong). The TIS (TIS = PS 9 IS) ranges from 0 to 12 with only nine possible values (that is, 0, 1, 2, 3, 4, 6, 8, 9 and 12). EpCAM ‘overexpression’ has been defined as a TIS [ 4 [17, 18]. Furthermore we defined four subgroups: no expression, TIS 0; weak expression, TIS 1–4; moderate expression, TIS 6 and 8; intense expression, TIS 9 and 12 [18]. Scores for Ki-67 were recorded as the number of immunopositive cells per high-power microscope (9400). The number of immunopositive cells under 5 microscopes per section with the highest cell counts was counted and the average was recorded. MVD was recorded as the number of vessels or clusters of cells immunopositive for CD34 per high-power microscope (9400). Each immunostained cell or cell cluster that was clearly separated from adjacent microvessels was considered as a single countable microvessel. MVD under 5 microscopes per section with the highest vascular counts were counted and the average was recorded.
Western blot analysis Total protein was extracted from 12 glioma samples (4 WHO grade II gliomas, 4 WHO grade III gliomas, and 4 WHO grade IV gliomas), using RIPA and 1 % PMSF (Biocolor Bioscience & Technology Company, Shanghai, China). The protein concentration of the samples was determined by the BCA assay (Biocolor Bioscience & Technology Company) according to the manufacturer’s protocol. Then equal amount of protein extracts were added to the 10 % SDS–polyacrylamide gels and electrophoresed. After transferred onto nitrocellulose membranes, proteins on the membranes were blocked with 5 % skim milk in Tris-saline buffer with 0.1 % Tween-20 (TBST), they were subsequently probed with primary antibodies at 4 °C overnight. After washings with TBST, secondary antibody conjugated with the horseradish peroxidase (Zhongshan Goldenbridge, Beijing, China) was added to the membranes. After washings with TBST, proteins were detected using the chemiluminescence detection kit (Millipore, Massachusmetts, USA). Antibodies used in this study included anti-EpCAM (ab71916, Abcam, Cambridge, UK, 1/1000), and anti-b-actin (AA128, Beyotime, Nantong, China; 1/5000). Western blot results were analyzed using the ImageJ2 9 Image software (National Institutes of Health, USA) and Multi Gauge Ver. 3.2 software (Fujifilm Life Science, Japan) by calculating the ratio of EpCAM integrated density (IntDen)/b-actin integrated density. Survival analysis Overall survival time was calculated from the date of diagnosis until death or the last follow-up. The patients who were out of contact (n = 6) or still alive (n = 8) at the endpoint of follow-up were referred to as ‘‘censored cases’’ (n = 14). The survival analysis of patients was calculated with the Kaplan–Meier method and the difference was analyzed using the two-sided log-rank test in SPSS 21.0 for windows. Cox Proportional-Hazards Model analysis in a stepwise method to evaluate the effect of multiple independent prognostic factors on survival outcome for all of 98 cases. Statistical analysis Statistical analysis was performed using SPSS 21.0 for windows. Chi square test (v2) and Independent-Samples Mann–Whitney U Test (U) were used to analyze immunohistochemical overexpression of EpCAM in glioma of different WHO classification. Independent sample t test (t) was used to analyze EpCAM expression difference (EpCAM/b-actin IntDen ratio) in Western Blot Analysis with Levene’s Test (F) for Equality of Variances.
123
J Neurooncol
Fig. 1 Representative immunohistochemical staining for EpCAM (4009). Membranous and Cytoplasmic staining of EpCAM was observed in (a–c); a WHO grade II malignant glioma with weak EpCAM expression(TIS = 4), slant arrow shows EpCAM staining on epithelial cell; b WHO grade III malignant glioma with moderate EpCAM expression(TIS = 8); c WHO grade IV with intense EpCAM
expression(TIS = 12). d intense membranous staining in intestine adenocarcinoma was showed as a positive control. Inserts show representative staining; Left-to-right arrows show membranous staining and right-to-left arrows show cytoplasmic staining.WHO, World Health Organization, EpCAM epithelial cell adhesion molecule, TIS total immunostaining score
Spearman rank correlation analysis was used to analyze the correlation of EpCAM expression with clinicopathological factors. Results are reported as being statistically significant if P \ 0.05 (2-sided).
gliomas, and grade III gliomas overexpression rate was higher compared to grade II gliomas (v2 = 5.467, P = 0.019; U = 394, P = 0.021) (Table 2). EpCAM overexpression correlated significantly with expression of Ki-67 (Spearman’s correlation coefficient, r = 0.475, P = 0.000), and MVD (r = 0.226, P = 0.026), but not with age(r = 0.174, P = 0.086) and gender (r = -0.161, P = 0.113) (Table 1). Western Blot analysis confirmed the results of immunochemistry, showing a stepwise increase in EpCAM expression from WHO II to WHO IV grade tumor samples (Fig. 2a). The differences of EpCAM/b-actin IntDen ratios are as follows, II versus III (t = -3.799, P = 0.009), II versus IV (t = -3.972, P = 0.028), and III versus IV (t = -2.508, P = 0.046) (Fig. 2b; Table 2). During the follow-up period, 84 of 98 glioma patients (85.7 %) had died [30 (81.1 %) from the no-EpCAM overexpression group and 45(86.5 %) from the EpCAM
Results Immunohistochemistry results showed that EpCAM was widely expressed in glioma (Table 1; Fig 1), which is 90.8 % (89 of 98), including weak expression (TIS 1–4) in 37 (37.8 %), moderate expression (TIS 6 and 8) in 20 (20.4 %) and intense expression (TIS 9 and 12) in 32 (32.7 %) samples (Table 1). The overexpression (moderate and intense expression) rate of WHO grade IV glioma was significantly higher than grade II (Chi square test, v2 = 20.984, P = 0.000; Mann–Whitney U Test, U = 922, P = 0.000) and III (v2 = 4.712, P = 0.030; U = 768.5, P = 0.031)
123
J Neurooncol Table 2 Discrepancy of EpCAM overexpression between gliomas of different WHO classifications WHO grades comparison
Immunohistochemistry v
2 a
(P*)
Western blot analysis a
U (P*)
Fb
tc
P*
II vs III
5.467 (0.019*)
394 (0.021*)
5.057
-3.799
0.009*
II vs IV
20.984 (0.000*)
922 (0.000*)
8.049
-3.972
0.028*
III vs IV
4.712 (0.030*)
768.5 (0.031*)
3.484
-2.508
0.046*
WHO World Health Organization Chi square test (v2) and Independent-Samples Mann–Whitney U Test(U) are used to test the immunohistochemical overexpression discrepancy between gliomas of different WHO grades
a
b
Levene’s Test (F) was used for Equality of Variances
c
Independent sample t test (t) was used to analyze EpCAM expression difference (EpCAM/b-actin IntDen ratio) in Western Blot Analysis between different WHO grades * P values indicating significance when P \ 0.05
Fig. 2 a Western blot analyses for EpCAM expression in malignant gliomas showed a stepwise intensive EpCAM expression in malignant gliomas from WHO II–IV grade. Due to the instruction manual of primary antibody (ab71916, Abcam, Cambridge, UK), the lower bands at 35 kDa are the targets for EpCAM. For a control, b-actin
expression levels were assessed in parallel. b EpCAM/b actin IntDen ratios were calculated to statistically show the expression differences. WHO World Health Organization, EpCAM epithelial cell adhesion molecule, IntDen integrated density
overexpression group]. The overall survival of WHO III and IV glioma patients with EpCAM overexpression was obviously lower than that without EpCAM overexpression (PIII = 0.041; PIV = 0.009) (Fig. 3; Table 3). Cox Proportional-Hazards Model analysis indicated that EpCAM overexpression (Risk ratio, RR = 2.470, P \ 0.01) is significant predictors for survival in the overall population of this study (Table 4).
Little is known about the exact signal-transducing role that EpCAM plays in normal and malignant cells until recently. In vitro studies in cancer cell lines have shown that blocking expression of EpCAM by short interfering RNAs could decrease proliferation, migration and invasion, suggesting a direct oncogenic role for EpCAM [15, 19]. Furthermore, EpCAM was found to directly up-regulate the oncogene c-myc [16] and play a role in the Wnt–b–catenin signaling [20–22], which is known to be involved in tumorigenesis and able to activate several target genes such as c-myc, vascular endothelial growth factor, and cyclooxygenase 2. The oncogenic function of EpCAM was confirmed in 2009, from the evidence that the positive influence of EpCAM on cell proliferation can be attributed to liberation of its ICD from the plasma membrane [22]. Furthermore, in animal experiments, the tumorigenic function of EpCAM was observed in a SCID mouse tumor model [22].
Discussion To our knowledge, this is the first study to show that EpCAM is widely expressed in glioma, which is 90.8 % (89 of 98), and the overexpression of EpCAM significantly correlates with glioma malignancy (WHO grades), MVD, proliferative factor Ki-67, and prognosis, but not with age and gender.
123
J Neurooncol
Fig. 3 Kaplan–Meier survival curves for malignant glioma patients (n = 98) with or without EpCAM overexpression. a Cumulative survival is significantly lower in total cases with EpCAM overexpression (P \ 0.01). b–d Subgroup analyses of prognostic value of EpCAM overexpression according to tumor WHO grade. The overall survival of patients with WHO grade III and IV glioma was
significantly worse with EpCAM overexpression group than without EpCAM overexpression group (PIII = 0.041; PIV = 0.009), but no significant difference was found for patients with WHO II grade. EpCAM epithelial cell adhesion molecule, Cum survival cumulative survival
Table 3 Kaplan–Meier survival analysis with log-rank test
Table 4 Cox multivariate analysis
EpCAM overexpression
Extent of resection
Chi square
P*
Chi square
P*
II
0.406
0.524
1.641
0.200
III
4.156
0.041*
0.004
0.953
IV Total
6.751
0.009*
1.291
0.256
16.697
0.000*
9.755
0.002*
* P values indicating significance when P \ 0.05
123
Risk ratio
P*
95 % confidence interval
Age
1.018
0.058
Adjuvant treatment
1.292
0.054
0.996–1.677
Extent of resection
1.508
0.083
0.948–2.399
EpCAM overexpression
2.470
\0.01*
1.561–3.909
* P values indicating significance when P \ 0.05
0.999–1.307
J Neurooncol
There has been a large database on EpCAM staining for cancers and normal tissues (including brain and brain tumors [23]) from its coming to light [11]. In this study, we used the anti-EpCAM antibody to demonstrate the overexpression of EpCAM in gliomas. Our results of EpCAM overexpression in glioma and its significant correlation with malignance (WHO grades) indicates that EpCAM is a potential molecular marker for glioma. That is to say, EpCAM can be used as a diagnostic marker to supplement the diagnostic criteria of WHO classification system and the detection of EpCAM expression may be a helpful tool for differentiation of glioma among different WHO degrees. For the past 4 decades, the expression of EpCAM has been investigated predominantly in malignant tissues for diagnostic and prognostic purposes [14, 15, 24], and a variety of therapeutic strategies have been specifically designed to target EpCAM based on its overexpression in carcinomas [25]. Furthermore, recently identified functions of EpCAM and EpICD in intracellular signaling transduction could greatly promote the future development of EpCAM-directed therapeutics [22]. In this study, the high overexpression rate in malignant glioma indicates that EpCAM can be a latent target for EpCAM-directed therapeutics, but more mechanism and prognosis-related researches should be taken into account to support this speculation. Tumor angiogenesis is referred to the formation of new vessels from preexisting vessels. In glioma, as with malignant tumors of other origins in the body, angiogenesis is necessary for tumor progression in the form of growth, invasion, and metastasis [26–28]. Among the frequentlyused angiogenesis factors, CD34 is well known as an endothelial marker and could give positive staining in physiologic and pathologic vessels [29, 30]. Combined with its good immunoreactivity, CD34 is considered an excellent marker for estimating MVD [29, 30]. This study shows that there is an significant correlation between EpCAM overexpression in gliomas and MVD (r = 0.226, P = 0.026), and sections of 66 patients (67.3 %) showed EpCAM staining on endothelial cells in immunohistochemistry (Fig. 1a). So, the findings in this study indicated that EpCAM may be involved in the angiogenesis of glioma, but the mechanism needs to be illuminated. Proliferation is a key feature of tumorigenesis and nucleus protein Ki-67, encoded by the MKI67 gene in humans [31], is an splendid marker for measuring the proliferative activity of tumor cells and strongly associated with the percentage of growth fraction in a given cell population, because of its existence in all active phases of the cell cycle (G(1), S, G(2), and mitosis) and absence from resting cells (G(0)) [32]. Expression of Ki-67 has been proved to correlate with tumorigenesis in brain tumors like gliomas and pituitary adenomas [33, 34]. This study shows
that EpCAM overexpression in glioma significantly correlates with Ki-67 expression (r = 0.475, P = 0.000), which implies that EpCAM may have a role in glioma proliferation. In the present study, age and gender have no significant correlation with EpCAM overexpression, which is in accord with previous results [35–37]. Kaplan–Meier analysis revealed that EpCAM overexpression correlates with the prognosis of glioma (grade II– IV, P \ 0.01), and in the subgroup, the survival time of patients with EpCAM overexpression was shorter than those of patients without EpCAM overexpression in malignant glioma (PIII = 0.041; PIV = 0.009). Cox Proportional-Hazards Model analysis confirmed that EpCAM overexpression (Risk ratio, RR = 2.470, P \ 0.01) was a reliable predictor of survival in the overall population of this study, which is similar to the observations in other lesions such as breast, ovarian, and esophageal cancers [12–14]. In our experience, the extent of resection was associated with improved survival for glioma patients. In a large-scale retrospective research, McGirt et al.[38] showed that increasing extent of resection was associated with improved survival for malignant gliomas. In this study, Kaplan–Meier univariate analysis showed that extent of resection was a significant predictor for survival (v2 = 9.755, P = 0.002)(Table 3). Limitations (1)
(2)
1 Due to the ethical and surgical principles, it is difficult to get enough normal brain tissues of human. So in this study we did not include normal glial tissues as experimental control. But based on the previous reports [23], it is assumed there is no EpCAM expression in healthy brain tissues. 2 Due to the limitations of laboratory conditions and time, we did not explore the downstream signal transduction of EpCAM in glioma.
Conclusions In summary, in spite of some limitations, our results convincingly showed a statistically significant stepwise increase in EpCAM overexpression from WHO II to IV grades of glioma and strong correlation with angiogenic factor MVD, proliferative factors Ki67 and prognosis. Consequently, our results indicate that EpCAM play a role in the tumorigenesis of human glioma and has the potential to be a diagnostic, prognostic and therapeutic molecular marker for human glioma. The downstream signaling transduction of EpCAM
123
J Neurooncol
in glioma is undetermined and need to be explored in the further studies.
16.
Acknowledgments This study was supported by the National Natural Science Foundation of China (Grant no. 81171062). The authors thank the patients who contributed to this research.
17.
Conflict of interest All authors declare that they have no conflict of interest.
18.
Ethical standards All the experiments in this article comply with the current laws of China. 19.
References 1. Ohgaki H (2009) Epidemiology of brain tumors. Methods Mol Biol 472:323–342. doi:10.1007/978-1-60327-492-0_14 2. Louis DN, Ohgaki H, Wiestler OD, Cavenee WK (2007) World Health Organization Classification of Tumours: Pathology and genetics of tumours of the nervous system. International Agency for Research on Cancer (IARC), Lyon 3. Wen PY, Kesari S (2008) Malignant gliomas in adults. N Engl J Med 359:492–507. doi:10.1056/NEJMra0708126 4. Stupp R, Mason WP, van den Bent MJ et al (2005) Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma. N Engl J Med 352:987–996 5. Rousseau A, Mokhtari K, Duyckaerts C (2008) The 2007 WHO classification of tumors of the central nervous system—what has changed? Curr Opin Neuro 21(6):720–727. doi:10.1097/WCO. 0b013e328312c3a7 6. Labussiere M, Wang XW, Idbaih A, Ducray F, Sanson M (2010) Prognostic markers in gliomas. Future Oncol 6(5):733–739. doi:10.2217/fon.10.25 7. Phillips HS, Kharbanda S, Chen R et al (2006) Molecular subclasses of high-grade glioma predict prognosis, delineate a pattern of disease progression, and resemble stages in neurogenesis. Cancer Cell 9(3):157–173. doi:10.1016/j.ccr.2006.02.019 8. Carpenter G, Red Brewer M (2009) EpCAM: another surface-tonucleus missile. Cancer Cell 15(3):165–166. doi:10.1016/j.ccr. 2009.02.005 9. Gires O, Klein CA, Baeuerle PA (2009) On the abundance of EpCAM on cancer stem cells. Nat Rev Cancer 9:143. doi:10. 1038/nrc2499-c1 10. Schnell U, Kuipers J, Giepmans BN (2013) EpCAM proteolysis: new fragments with distinct functions? Biosci Rep 33(2):e00030. doi:10.1042/BSR20120128 11. Herlyn D, Herlyn M, Steplewski Z, Koprowski H (1979) Monoclonal antibodies in cell-mediated cytotoxicity against human melanoma and colorectal carcinoma. Eur J Immunol 9:657–659. doi:10.1002/eji.1830090817 12. Baeuerle PA, Gires O (2007) EpCAM (CD326) finding its role in cancer. Br J Cancer 96:417–423. doi:10.1038/sj.bjc.6603494 13. van der Gun BT, Melchers LJ, Ruiters MH, de Leij LF, McLaughlin PM, Rots MG (2010) EpCAM in carcinogenesis: the good, the bad or the ugly. Carcinogenesis 31(11):1913–1921. doi:10.1093/carcin/bgq187 14. Patriarca C, Macchi RM, Marschner AK, Mellstedt H (2012) Epithelial cell adhesion molecule expression (CD326) in cancer: a short review. Cancer Treat Rev 38(1):68–75. doi:10.1016/j.ctrv. 2011.04.002 15. Osta WA, Chen Y, Mikhitarian K et al (2004) EpCAM is overexpressed in breast cancer and is a potential target for breast
123
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
cancer gene therapy. Cancer Res 64:5818–5824. doi:10.1158/ 0008-5472.CAN-04-0754 Mu¨nz M, Kieu C, Mack B, Schmitt B, Zeidler R, Gires O (2004) The carcinoma-associated antigen EpCAM upregulates c-myc and induces cell proliferation. Oncogene 23:5748–5758. doi:10. 1038/sj.onc.1207610 Gastl G, Spizzo G, Obrist P, Du¨nser M, Mikuz G (2000) EpCAM overexpression in breast cancer as a predictor of survival. Lancet 356:1981–1982. doi:10.1016/S0140-6736(00)03312-2 Spizzo G, Fong D, Wurm M et al (2011) EpCAM expression in primary tumour tissues and metastases: an immunohistochemical analysis. J Clin Pathol 64(5):415–420. doi:10.1136/jcp.2011. 090274 Yanamoto S, Kawasaki G, Yoshitomi I, Iwamoto T, Hirata K, Mizuno A (2007) Clinicopathologic significance of EpCAM expression in squamous cell carcinoma of the tongue and its possibility as a potential target for tongue cancer gene therapy. Oral Oncol 43:869–877. doi:10.1016/j.oraloncology.2006.10.010 Yamashita T, Budhu A, Forgues M, Wang XW (2007) Activation of Hepatic Stem Cell Marker EpCAM by Wnt Signaling in Hepatocellular Carcinoma. Cancer Res 67(22):10831–10839. doi:10.1158/0008-5472 Imrich S, Hachmeister M, Gires O (2012) EpCAM and its potential role in tumor-initiating cells. Cell Adh Migr 6(1):30–38. doi:10.4161/cam.18953 Maetzel D, Denzel S, Mack B et al (2009) Nuclear signalling by tumour-associated antigen EpCAM. Nat Cell Biol 11(2): 162–171. doi:10.1038/ncb1824 Went PT, Lugli A, Meier S et al (2004) Frequent EpCam protein expression in human carcinomas. Hum Pathol 35(1):122–128. doi:10.1016/j.humpath.2003.08.026 Subira´ D, Serrano C, Castan˜o´n S et al (2012) Role of flow cytometry immunophenotyping in the diagnosis of leptomeningeal carcinomatosis. Neuro Oncol 14(1):43–52. doi:10.1093/ neuonc/nor172 Kurtz JE, Dufour P (2010) Adecatumumab: an anti-EpCAM monoclonal antibody, from the bench to the bedside. Expert Opin Biol Ther 10(6):951–958. doi:10.1517/14712598.2010.482098 Chaudhry IH, O’Donovan DG, Brenchley PE, Reid H, Roberts IS (2001) Vascular endothelial growth factor expression correlates with tumour grade and vascularity in gliomas. Histopathology 39:409–415. doi:10.1046/j.1365-2559.2001.01230.x Rahmah NN, Sakai K, Sano K, Hongo K (2012) Expression of RECK in endothelial cells of glioma: comparison with CD34 and VEGF expressions. J Neurooncol 107(3):559–564. doi:10.1007/ s11060-011-0778-z Netto GC, Bleil CB, Hilbig A, Coutinho LM (2008) Immunohistochemical evaluation of the microvascular density through the expression of TGF-beta (CD 105/endoglin) and CD 34 receptors and expression of the vascular endothelial growth factor (VEGF) in oligodendrogliomas. Neuropathology 28:17–23. doi:10.1111/j.1440-1789.2007.00825.x Vermeulen PB, Gasparini G, Fox SB et al (1996) Quantification of angiogenesis in solid human tumours: an international consensus on the methodology and criteria of evaluation. Eur J Cancer 32A:2474–2484 Deb P, Boruah D, Dutta V (2012) Morphometric study of microvessels in primary CNS tumors and its correlation with tumor types and grade. Microvasc Res 84(1):34–43. doi:10.1016/ j.mvr.2012.03.004 Varga Z, Diebold J, Dommann-Scherrer C et al (2012) How reliable is Ki-67 immunohistochemistry in grade 2 breast carcinomas? A QA study of the Swiss Working Group of Breast-and Gynecopathologists. PLoS One 7:e37379. doi:10.1371/journal. pone.0037379
J Neurooncol 32. Scholzen T, Gerdes J (2000) The Ki-67 protein: from the known and the unknown. J Cell Physiol 182:311–322. doi:10.1002/ (SICI)1097-4652(200003)182:3\311:AID-JCP1[3.0.CO;2-9 33. Hou XZ, Liu W, Fan HT et al (2010) Expression of hepatocyte growth factor and its receptor c-Met in human pituitary adenomas. Neuro Oncol 12(8):799–803. doi:10.1093/neuonc/noq024 34. Pang B, Fan H, Zhang IY et al (2012) HMGA1 expression in human gliomas and its correlation with tumor proliferation, invasion and angiogenesis. J Neurooncol 106(3):543–549. doi:10. 1007/s11060-011-0710-6 35. Fong D, Steurer M, Obrist P et al (2008) Ep-CAM expression in pancreatic and ampullary carcinomas: frequency and prognostic relevance. J Clin Pathol 61(1):31–35. doi:10.1136/jcp.2006.037333
36. Varga Meri, Obrist Peter, Schneeberger Stephan et al (2004) Overexpression of epithelial cell adhesion molecule antigen in gallbladder carcinoma is an independent marker for poor survival. Clin Cancer Res 10:3131–3136. doi:10.1158/1078-0432 37. Spizzo G, Went P, Dirnhofer S, Obrist P et al (2006) Overexpression of epithelial cell adhesion molecule (Ep-CAM) is an independent prognostic marker for reduced survival of patients with epithelial ovarian cancer. Gynecol Oncol 103(2):483–488. doi:10.1016/j.ygyno.2006.03.035 38. McGirt MJ, Chaichana KL, Gathinji M et al (2009) Independent association of extent of resection with survival in patients with malignant brain astrocytoma. J Neurosurg 110(1):156–162. doi:10.3171/2008.4.17536
123
LABORATORY INVESTIGATION
The overexpression of Epithelial cell adhesion molecule (EpCAM) in glioma Xin Chen • Wei-Yuan Ma • Shang-Chen Xu • Yu Liang • Yi-Bing Fu Bo Pang • Tao Xin • Hai-Tao Fan • Rui Zhang • Jian-Gang Luo • Wen-Qing Kang • Min Wang • Qi Pang
•
Received: 6 August 2013 / Accepted: 23 April 2014 Ó Springer Science+Business Media New York 2014
Abstract Epithelial cell adhesion molecule (EpCAM) is overexpressed in various neoplasms as a tumor-associated antigen and absent in natural brain. However, little is known about EpCAM’s expression in gliomas. To investigate the expression of EpCAM in gliomas and understand the correlation of EpCAM expression with malignancy, proliferation, angiogenesis, and prognosis, we studied the expression of EpCAM in 98 glioma samples by immunohistochemistry and by western blotting (N = 12). Correlative analysis of EpCAM overexpression with microvessel density (MVD), Ki-67 expression, age, and gender were made. Survival data was analyzed with Kaplan–Meier method and Cox Proportional Hazard Model. Immunohistochemistry results showed EpCAM was widely expressed in glioma (90.8 %). The overexpression rate of WHO grade IV gliomas was significantly higher EpCAM
overexpression correlated significantly with Ki-67 expression and MVD. Western blot analysis also revealed a stepwise increase in EpCAM expression from WHO II to IV glioma. The overall survival of WHO III and IV glioma patients with EpCAM overexpression was obviously lower than that without EpCAM overexpression. EpCAM overexpression was an independent prognostic factor for overall survival in glioma patients. This study firstly shows that EpCAM overexpression correlates significantly with malignancy (WHO grades), proliferation (Ki67), angiogenesis (MVD), and prognosis in gliomas. EpCAM may participate in tumorgenesis of gliomas. Keywords Glioma Epithelial cell adhesion molecule (EpCAM) Overexpression Tumorigenesis
Introduction X. Chen S.-C. Xu T. Xin H.-T. Fan R. Zhang Q. Pang (&) Department of Neurosurgery, Shandong Provincial Hospital Affiliated to Shandong University, Jinan 250021, People’s Republic of China e-mail: [email protected] X. Chen W.-Y. Ma Y. Liang B. Pang R. Zhang J.-G. Luo W.-Q. Kang Shandong University School of Medicine, Jinan 250012, People’s Republic of China Y.-B. Fu Department of Gynaecology, Shandong Provincial Hospital Affiliated to Shandong University, Jinan 250021, People’s Republic of China M. Wang (&) Department of Neurology, Yantai Mountain Hospital, Yantai 264001, People’s Republic of China e-mail: [email protected]
Glioma, the most common neoplasm of central nervous system, accounts for more than 70 % of brain tumors [1]. According to the World Health Organization (WHO) classification, the most used glioma grading system [2], gliomas are subdivided to four grades (I–IV) based on the degree of malignancy. According to the WHO grades, I and II gliomas are generally defined to be benign, and gliomas of higher grades (III–IV), referred to malignant glioma, is more invasive and has worse outcomes [3, 4]. Although the WHO classification for glioma is accepted widely, but it is far from ideal [5]. Recent studies have identified new diagnostic and prognostic markers to supplement the WHO histological classification, such as the 1p19q co-deletion, O6-methylguanine DNA methyltransferase status, and mutations of isocitrate dehydrogenases 1 and 2 [6]. However, the understanding about
123
J Neurooncol
the pathogenesis of glioma is still largely unclear and the prognosis of patients with malignant glioma is still poor despite optimal treatments which combine surgery, chemotherapy, and irradiation [1, 3, 6, 7]. There thus is an urgent need to identify novel molecular markers for diagnosis and as potential therapeutic targets for glioma. Epithelial cell adhesion molecule (EpCAM) is a glycosylated, type I transmembrane protein, which is expressed in human cancers, progenitor and stem cells [8–10]. EpCAM was initially reported as a dominant surface antigen on human colon carcinoma [11], further comprehensive analysis of EpCAM expression has shown that EpCAM is expressed in many human cancers [12]. Because of the tumor-specific expression of EpCAM, it has been explored as a prognostic/diagnostic marker and anti-cancer target for immunotherapeutic strategies since 1970s [12–14]. However, it was not until the twentieth century that the roles of EpCAM in the intracellular signaling pathways for proliferation, migration, and mitogenic signal transduction were revealed [15, 16]. We sought to understand the role of EpCAM in tumorigenesis of glioma. The aim was to investigate the relationship of WHO classification of gliomas with some clinicopathologic factors, and prognosis. The results showed that EpCAM is widely expressed in gliomas and revealed a statistically significant correlation of EpCAM overexpression with the WHO classification, proliferation factor ki-67, microvessel density (MVD), and prognosis.
Table 1 Clinicopathologic characteristics of patients and Immunohistochemical expression of EpCAM in gliomas with different WHO classification Variables
III (25)
IV (49)
Total (98)
No expression(TIS 0)
4
3
2
9
Weak expression(TIS 1–4)
16
10
11
37
Moderate expression(TIS 6,8)
3
7
10
20
Intense expression(TIS 9,12)
1
5
26
32
Male
8
14
29
51
Female
16
11
20
47
P*
-0.161
0.113
NO overexpression
Overexpression
Gender
Extent of resection [90 %
18
14
20
52
\90 %
6
11
29
46
None
1
1
2
4
Radiotherapy only
18
7
6
31
Chemotherapy only
0
0
2
2
Radiotherapy and chemotherapy combination
5
17
39
61
Adjuvant treatment
Patients and samples
123
II (24)
ra
EpCAM overexpression
Materials and methods
We analyzed the archival records from the collection of glioma files of Provincial Hospital Affiliated to Shandong University, Jinan, P. R. China, and selected 98 cases from consecutive patients according to the criteria:1 Undergoing surgeries from 2003 to 2013; 2 Agreeing to contribute to this study and signing consents; 3 Preoperative Karnofsky performance score score [70; 4 None of the patients had received chemotherapy or radiotherapy prior to surgery. The Clinical data (including grade, age, gender and extent of resection) were obtained from patients’ medical records or attending physicians (Table 1). Contacts, adjacent treatment, and survival information were obtained from hospital information systems and the patients or patients’ relatives. Sections of all tumors were produced from paraffin-embedded tissue samples stored in Department of Pathology and reviewed by 2 pathologists to define the histological grade. 12 fresh glioma specimens were collected and stored by the Institute of Neurosurgery in Provincial Hospital Affiliated to Shandong University and
WHO grades (number)
Age(x)
38.2
47.0
51.8
47.3
0.174
0.086
Ki-67(x)
7.6
23.9
39.9
27.9
0.475
0.000*
MVD(x)
34.6
44.1
54.3
46.8
0.226
0.026*
WHO World Health Organization, MVD microvessel density, x mean value, EpCAM epithelial cell adhesion molecule, TIS total immunostaining score a
Spearman rank correlation coefficient (r) is used to analyze the correlation of EpCAM with clinicopathological parameters * P values indicating significance when P \ 0.05
diagnosed by Departments of Pathology. In this study, all the operations have been performed in accordance with the ethical standards and all the experiments comply with the current laws of China. Immunohistochemistry The immunohistochemical study was performed using the Envision PV-style two-step method (PV-9000 Polymer Detection System, Zhongshan Goldenbridge Biotechnology, China). A section of intestinal adenocarcinoma with known EpCAM positivity [12] was used as a positive control. For the negative control, primary antibodies were
J Neurooncol
replaced by the preparation solution for antibodies. Formalin-fixed, paraffin-embedded tissue sections of 4-lM thickness were baked at 60 °C for 30 min, deparaffinized in xylene, and rehydrated in graded concentrations of ethanol. Heat-induced antigen retrieval (10 mM citrate buffer [pH 6.0] at 98 °C for 20 min in a thermostatcontrolled waterbath) was carried out. Endogenous peroxidase activity was quenched by incubating in 0.3 % hydrogen peroxide at 37 °C for 30 min. Nonspecific staining was blocked by normal serum from the same species as that of secondary antibodies at 37 °C for 30 min. Primary antibodies (anti-EpCAM, ab71916, Abcam, Cambridge, UK, 1/100; anti-Ki67, ab66155, Abcam, Cambridge, UK, 1/200; anti-CD34, ab110643, Abcam, Cambridge, UK, 1/200) were applied at 4 °C overnight in a humidified chamber, then rewarming at 37 °C for 30 min, followed by Polymer Helper (Zhongshan Goldenbridge Biotechnology) incubation at 37 °C for 20 min and polyperoxidase-anti-mouse/rabbit IgG (Zhongshan Goldenbridge Biotechnology) incubation at 37 °C for 25 min. Diaminobenzidine was used as the enzyme substrate to observe the specific antibody localization, and hematoxylin was used as a nuclear counterstain. Sections were examined and scored for EpCAM, MVD and Ki-67 by two observers who were unaware of the histological diagnoses or clinical features. All the samples were stained at the same time. EpCAM expression for glioma staining was evaluated by calculating a total immunostaining score (TIS) as the product of a proportion score (PS) and an intensity score (IS). The PS describes the estimated fraction of positively stained cells (0, none; 1, \10 %; 2, 10–50 %; 3, 51–80 %; 4, [80 %). The IS represents the estimated staining intensity as compared to the controls (0 no staining; 1 weak; 2 moderate; 3 strong). The TIS (TIS = PS 9 IS) ranges from 0 to 12 with only nine possible values (that is, 0, 1, 2, 3, 4, 6, 8, 9 and 12). EpCAM ‘overexpression’ has been defined as a TIS [ 4 [17, 18]. Furthermore we defined four subgroups: no expression, TIS 0; weak expression, TIS 1–4; moderate expression, TIS 6 and 8; intense expression, TIS 9 and 12 [18]. Scores for Ki-67 were recorded as the number of immunopositive cells per high-power microscope (9400). The number of immunopositive cells under 5 microscopes per section with the highest cell counts was counted and the average was recorded. MVD was recorded as the number of vessels or clusters of cells immunopositive for CD34 per high-power microscope (9400). Each immunostained cell or cell cluster that was clearly separated from adjacent microvessels was considered as a single countable microvessel. MVD under 5 microscopes per section with the highest vascular counts were counted and the average was recorded.
Western blot analysis Total protein was extracted from 12 glioma samples (4 WHO grade II gliomas, 4 WHO grade III gliomas, and 4 WHO grade IV gliomas), using RIPA and 1 % PMSF (Biocolor Bioscience & Technology Company, Shanghai, China). The protein concentration of the samples was determined by the BCA assay (Biocolor Bioscience & Technology Company) according to the manufacturer’s protocol. Then equal amount of protein extracts were added to the 10 % SDS–polyacrylamide gels and electrophoresed. After transferred onto nitrocellulose membranes, proteins on the membranes were blocked with 5 % skim milk in Tris-saline buffer with 0.1 % Tween-20 (TBST), they were subsequently probed with primary antibodies at 4 °C overnight. After washings with TBST, secondary antibody conjugated with the horseradish peroxidase (Zhongshan Goldenbridge, Beijing, China) was added to the membranes. After washings with TBST, proteins were detected using the chemiluminescence detection kit (Millipore, Massachusmetts, USA). Antibodies used in this study included anti-EpCAM (ab71916, Abcam, Cambridge, UK, 1/1000), and anti-b-actin (AA128, Beyotime, Nantong, China; 1/5000). Western blot results were analyzed using the ImageJ2 9 Image software (National Institutes of Health, USA) and Multi Gauge Ver. 3.2 software (Fujifilm Life Science, Japan) by calculating the ratio of EpCAM integrated density (IntDen)/b-actin integrated density. Survival analysis Overall survival time was calculated from the date of diagnosis until death or the last follow-up. The patients who were out of contact (n = 6) or still alive (n = 8) at the endpoint of follow-up were referred to as ‘‘censored cases’’ (n = 14). The survival analysis of patients was calculated with the Kaplan–Meier method and the difference was analyzed using the two-sided log-rank test in SPSS 21.0 for windows. Cox Proportional-Hazards Model analysis in a stepwise method to evaluate the effect of multiple independent prognostic factors on survival outcome for all of 98 cases. Statistical analysis Statistical analysis was performed using SPSS 21.0 for windows. Chi square test (v2) and Independent-Samples Mann–Whitney U Test (U) were used to analyze immunohistochemical overexpression of EpCAM in glioma of different WHO classification. Independent sample t test (t) was used to analyze EpCAM expression difference (EpCAM/b-actin IntDen ratio) in Western Blot Analysis with Levene’s Test (F) for Equality of Variances.
123
J Neurooncol
Fig. 1 Representative immunohistochemical staining for EpCAM (4009). Membranous and Cytoplasmic staining of EpCAM was observed in (a–c); a WHO grade II malignant glioma with weak EpCAM expression(TIS = 4), slant arrow shows EpCAM staining on epithelial cell; b WHO grade III malignant glioma with moderate EpCAM expression(TIS = 8); c WHO grade IV with intense EpCAM
expression(TIS = 12). d intense membranous staining in intestine adenocarcinoma was showed as a positive control. Inserts show representative staining; Left-to-right arrows show membranous staining and right-to-left arrows show cytoplasmic staining.WHO, World Health Organization, EpCAM epithelial cell adhesion molecule, TIS total immunostaining score
Spearman rank correlation analysis was used to analyze the correlation of EpCAM expression with clinicopathological factors. Results are reported as being statistically significant if P \ 0.05 (2-sided).
gliomas, and grade III gliomas overexpression rate was higher compared to grade II gliomas (v2 = 5.467, P = 0.019; U = 394, P = 0.021) (Table 2). EpCAM overexpression correlated significantly with expression of Ki-67 (Spearman’s correlation coefficient, r = 0.475, P = 0.000), and MVD (r = 0.226, P = 0.026), but not with age(r = 0.174, P = 0.086) and gender (r = -0.161, P = 0.113) (Table 1). Western Blot analysis confirmed the results of immunochemistry, showing a stepwise increase in EpCAM expression from WHO II to WHO IV grade tumor samples (Fig. 2a). The differences of EpCAM/b-actin IntDen ratios are as follows, II versus III (t = -3.799, P = 0.009), II versus IV (t = -3.972, P = 0.028), and III versus IV (t = -2.508, P = 0.046) (Fig. 2b; Table 2). During the follow-up period, 84 of 98 glioma patients (85.7 %) had died [30 (81.1 %) from the no-EpCAM overexpression group and 45(86.5 %) from the EpCAM
Results Immunohistochemistry results showed that EpCAM was widely expressed in glioma (Table 1; Fig 1), which is 90.8 % (89 of 98), including weak expression (TIS 1–4) in 37 (37.8 %), moderate expression (TIS 6 and 8) in 20 (20.4 %) and intense expression (TIS 9 and 12) in 32 (32.7 %) samples (Table 1). The overexpression (moderate and intense expression) rate of WHO grade IV glioma was significantly higher than grade II (Chi square test, v2 = 20.984, P = 0.000; Mann–Whitney U Test, U = 922, P = 0.000) and III (v2 = 4.712, P = 0.030; U = 768.5, P = 0.031)
123
J Neurooncol Table 2 Discrepancy of EpCAM overexpression between gliomas of different WHO classifications WHO grades comparison
Immunohistochemistry v
2 a
(P*)
Western blot analysis a
U (P*)
Fb
tc
P*
II vs III
5.467 (0.019*)
394 (0.021*)
5.057
-3.799
0.009*
II vs IV
20.984 (0.000*)
922 (0.000*)
8.049
-3.972
0.028*
III vs IV
4.712 (0.030*)
768.5 (0.031*)
3.484
-2.508
0.046*
WHO World Health Organization Chi square test (v2) and Independent-Samples Mann–Whitney U Test(U) are used to test the immunohistochemical overexpression discrepancy between gliomas of different WHO grades
a
b
Levene’s Test (F) was used for Equality of Variances
c
Independent sample t test (t) was used to analyze EpCAM expression difference (EpCAM/b-actin IntDen ratio) in Western Blot Analysis between different WHO grades * P values indicating significance when P \ 0.05
Fig. 2 a Western blot analyses for EpCAM expression in malignant gliomas showed a stepwise intensive EpCAM expression in malignant gliomas from WHO II–IV grade. Due to the instruction manual of primary antibody (ab71916, Abcam, Cambridge, UK), the lower bands at 35 kDa are the targets for EpCAM. For a control, b-actin
expression levels were assessed in parallel. b EpCAM/b actin IntDen ratios were calculated to statistically show the expression differences. WHO World Health Organization, EpCAM epithelial cell adhesion molecule, IntDen integrated density
overexpression group]. The overall survival of WHO III and IV glioma patients with EpCAM overexpression was obviously lower than that without EpCAM overexpression (PIII = 0.041; PIV = 0.009) (Fig. 3; Table 3). Cox Proportional-Hazards Model analysis indicated that EpCAM overexpression (Risk ratio, RR = 2.470, P \ 0.01) is significant predictors for survival in the overall population of this study (Table 4).
Little is known about the exact signal-transducing role that EpCAM plays in normal and malignant cells until recently. In vitro studies in cancer cell lines have shown that blocking expression of EpCAM by short interfering RNAs could decrease proliferation, migration and invasion, suggesting a direct oncogenic role for EpCAM [15, 19]. Furthermore, EpCAM was found to directly up-regulate the oncogene c-myc [16] and play a role in the Wnt–b–catenin signaling [20–22], which is known to be involved in tumorigenesis and able to activate several target genes such as c-myc, vascular endothelial growth factor, and cyclooxygenase 2. The oncogenic function of EpCAM was confirmed in 2009, from the evidence that the positive influence of EpCAM on cell proliferation can be attributed to liberation of its ICD from the plasma membrane [22]. Furthermore, in animal experiments, the tumorigenic function of EpCAM was observed in a SCID mouse tumor model [22].
Discussion To our knowledge, this is the first study to show that EpCAM is widely expressed in glioma, which is 90.8 % (89 of 98), and the overexpression of EpCAM significantly correlates with glioma malignancy (WHO grades), MVD, proliferative factor Ki-67, and prognosis, but not with age and gender.
123
J Neurooncol
Fig. 3 Kaplan–Meier survival curves for malignant glioma patients (n = 98) with or without EpCAM overexpression. a Cumulative survival is significantly lower in total cases with EpCAM overexpression (P \ 0.01). b–d Subgroup analyses of prognostic value of EpCAM overexpression according to tumor WHO grade. The overall survival of patients with WHO grade III and IV glioma was
significantly worse with EpCAM overexpression group than without EpCAM overexpression group (PIII = 0.041; PIV = 0.009), but no significant difference was found for patients with WHO II grade. EpCAM epithelial cell adhesion molecule, Cum survival cumulative survival
Table 3 Kaplan–Meier survival analysis with log-rank test
Table 4 Cox multivariate analysis
EpCAM overexpression
Extent of resection
Chi square
P*
Chi square
P*
II
0.406
0.524
1.641
0.200
III
4.156
0.041*
0.004
0.953
IV Total
6.751
0.009*
1.291
0.256
16.697
0.000*
9.755
0.002*
* P values indicating significance when P \ 0.05
123
Risk ratio
P*
95 % confidence interval
Age
1.018
0.058
Adjuvant treatment
1.292
0.054
0.996–1.677
Extent of resection
1.508
0.083
0.948–2.399
EpCAM overexpression
2.470
\0.01*
1.561–3.909
* P values indicating significance when P \ 0.05
0.999–1.307
J Neurooncol
There has been a large database on EpCAM staining for cancers and normal tissues (including brain and brain tumors [23]) from its coming to light [11]. In this study, we used the anti-EpCAM antibody to demonstrate the overexpression of EpCAM in gliomas. Our results of EpCAM overexpression in glioma and its significant correlation with malignance (WHO grades) indicates that EpCAM is a potential molecular marker for glioma. That is to say, EpCAM can be used as a diagnostic marker to supplement the diagnostic criteria of WHO classification system and the detection of EpCAM expression may be a helpful tool for differentiation of glioma among different WHO degrees. For the past 4 decades, the expression of EpCAM has been investigated predominantly in malignant tissues for diagnostic and prognostic purposes [14, 15, 24], and a variety of therapeutic strategies have been specifically designed to target EpCAM based on its overexpression in carcinomas [25]. Furthermore, recently identified functions of EpCAM and EpICD in intracellular signaling transduction could greatly promote the future development of EpCAM-directed therapeutics [22]. In this study, the high overexpression rate in malignant glioma indicates that EpCAM can be a latent target for EpCAM-directed therapeutics, but more mechanism and prognosis-related researches should be taken into account to support this speculation. Tumor angiogenesis is referred to the formation of new vessels from preexisting vessels. In glioma, as with malignant tumors of other origins in the body, angiogenesis is necessary for tumor progression in the form of growth, invasion, and metastasis [26–28]. Among the frequentlyused angiogenesis factors, CD34 is well known as an endothelial marker and could give positive staining in physiologic and pathologic vessels [29, 30]. Combined with its good immunoreactivity, CD34 is considered an excellent marker for estimating MVD [29, 30]. This study shows that there is an significant correlation between EpCAM overexpression in gliomas and MVD (r = 0.226, P = 0.026), and sections of 66 patients (67.3 %) showed EpCAM staining on endothelial cells in immunohistochemistry (Fig. 1a). So, the findings in this study indicated that EpCAM may be involved in the angiogenesis of glioma, but the mechanism needs to be illuminated. Proliferation is a key feature of tumorigenesis and nucleus protein Ki-67, encoded by the MKI67 gene in humans [31], is an splendid marker for measuring the proliferative activity of tumor cells and strongly associated with the percentage of growth fraction in a given cell population, because of its existence in all active phases of the cell cycle (G(1), S, G(2), and mitosis) and absence from resting cells (G(0)) [32]. Expression of Ki-67 has been proved to correlate with tumorigenesis in brain tumors like gliomas and pituitary adenomas [33, 34]. This study shows
that EpCAM overexpression in glioma significantly correlates with Ki-67 expression (r = 0.475, P = 0.000), which implies that EpCAM may have a role in glioma proliferation. In the present study, age and gender have no significant correlation with EpCAM overexpression, which is in accord with previous results [35–37]. Kaplan–Meier analysis revealed that EpCAM overexpression correlates with the prognosis of glioma (grade II– IV, P \ 0.01), and in the subgroup, the survival time of patients with EpCAM overexpression was shorter than those of patients without EpCAM overexpression in malignant glioma (PIII = 0.041; PIV = 0.009). Cox Proportional-Hazards Model analysis confirmed that EpCAM overexpression (Risk ratio, RR = 2.470, P \ 0.01) was a reliable predictor of survival in the overall population of this study, which is similar to the observations in other lesions such as breast, ovarian, and esophageal cancers [12–14]. In our experience, the extent of resection was associated with improved survival for glioma patients. In a large-scale retrospective research, McGirt et al.[38] showed that increasing extent of resection was associated with improved survival for malignant gliomas. In this study, Kaplan–Meier univariate analysis showed that extent of resection was a significant predictor for survival (v2 = 9.755, P = 0.002)(Table 3). Limitations (1)
(2)
1 Due to the ethical and surgical principles, it is difficult to get enough normal brain tissues of human. So in this study we did not include normal glial tissues as experimental control. But based on the previous reports [23], it is assumed there is no EpCAM expression in healthy brain tissues. 2 Due to the limitations of laboratory conditions and time, we did not explore the downstream signal transduction of EpCAM in glioma.
Conclusions In summary, in spite of some limitations, our results convincingly showed a statistically significant stepwise increase in EpCAM overexpression from WHO II to IV grades of glioma and strong correlation with angiogenic factor MVD, proliferative factors Ki67 and prognosis. Consequently, our results indicate that EpCAM play a role in the tumorigenesis of human glioma and has the potential to be a diagnostic, prognostic and therapeutic molecular marker for human glioma. The downstream signaling transduction of EpCAM
123
J Neurooncol
in glioma is undetermined and need to be explored in the further studies.
16.
Acknowledgments This study was supported by the National Natural Science Foundation of China (Grant no. 81171062). The authors thank the patients who contributed to this research.
17.
Conflict of interest All authors declare that they have no conflict of interest.
18.
Ethical standards All the experiments in this article comply with the current laws of China. 19.
References 1. Ohgaki H (2009) Epidemiology of brain tumors. Methods Mol Biol 472:323–342. doi:10.1007/978-1-60327-492-0_14 2. Louis DN, Ohgaki H, Wiestler OD, Cavenee WK (2007) World Health Organization Classification of Tumours: Pathology and genetics of tumours of the nervous system. International Agency for Research on Cancer (IARC), Lyon 3. Wen PY, Kesari S (2008) Malignant gliomas in adults. N Engl J Med 359:492–507. doi:10.1056/NEJMra0708126 4. Stupp R, Mason WP, van den Bent MJ et al (2005) Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma. N Engl J Med 352:987–996 5. Rousseau A, Mokhtari K, Duyckaerts C (2008) The 2007 WHO classification of tumors of the central nervous system—what has changed? Curr Opin Neuro 21(6):720–727. doi:10.1097/WCO. 0b013e328312c3a7 6. Labussiere M, Wang XW, Idbaih A, Ducray F, Sanson M (2010) Prognostic markers in gliomas. Future Oncol 6(5):733–739. doi:10.2217/fon.10.25 7. Phillips HS, Kharbanda S, Chen R et al (2006) Molecular subclasses of high-grade glioma predict prognosis, delineate a pattern of disease progression, and resemble stages in neurogenesis. Cancer Cell 9(3):157–173. doi:10.1016/j.ccr.2006.02.019 8. Carpenter G, Red Brewer M (2009) EpCAM: another surface-tonucleus missile. Cancer Cell 15(3):165–166. doi:10.1016/j.ccr. 2009.02.005 9. Gires O, Klein CA, Baeuerle PA (2009) On the abundance of EpCAM on cancer stem cells. Nat Rev Cancer 9:143. doi:10. 1038/nrc2499-c1 10. Schnell U, Kuipers J, Giepmans BN (2013) EpCAM proteolysis: new fragments with distinct functions? Biosci Rep 33(2):e00030. doi:10.1042/BSR20120128 11. Herlyn D, Herlyn M, Steplewski Z, Koprowski H (1979) Monoclonal antibodies in cell-mediated cytotoxicity against human melanoma and colorectal carcinoma. Eur J Immunol 9:657–659. doi:10.1002/eji.1830090817 12. Baeuerle PA, Gires O (2007) EpCAM (CD326) finding its role in cancer. Br J Cancer 96:417–423. doi:10.1038/sj.bjc.6603494 13. van der Gun BT, Melchers LJ, Ruiters MH, de Leij LF, McLaughlin PM, Rots MG (2010) EpCAM in carcinogenesis: the good, the bad or the ugly. Carcinogenesis 31(11):1913–1921. doi:10.1093/carcin/bgq187 14. Patriarca C, Macchi RM, Marschner AK, Mellstedt H (2012) Epithelial cell adhesion molecule expression (CD326) in cancer: a short review. Cancer Treat Rev 38(1):68–75. doi:10.1016/j.ctrv. 2011.04.002 15. Osta WA, Chen Y, Mikhitarian K et al (2004) EpCAM is overexpressed in breast cancer and is a potential target for breast
123
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
cancer gene therapy. Cancer Res 64:5818–5824. doi:10.1158/ 0008-5472.CAN-04-0754 Mu¨nz M, Kieu C, Mack B, Schmitt B, Zeidler R, Gires O (2004) The carcinoma-associated antigen EpCAM upregulates c-myc and induces cell proliferation. Oncogene 23:5748–5758. doi:10. 1038/sj.onc.1207610 Gastl G, Spizzo G, Obrist P, Du¨nser M, Mikuz G (2000) EpCAM overexpression in breast cancer as a predictor of survival. Lancet 356:1981–1982. doi:10.1016/S0140-6736(00)03312-2 Spizzo G, Fong D, Wurm M et al (2011) EpCAM expression in primary tumour tissues and metastases: an immunohistochemical analysis. J Clin Pathol 64(5):415–420. doi:10.1136/jcp.2011. 090274 Yanamoto S, Kawasaki G, Yoshitomi I, Iwamoto T, Hirata K, Mizuno A (2007) Clinicopathologic significance of EpCAM expression in squamous cell carcinoma of the tongue and its possibility as a potential target for tongue cancer gene therapy. Oral Oncol 43:869–877. doi:10.1016/j.oraloncology.2006.10.010 Yamashita T, Budhu A, Forgues M, Wang XW (2007) Activation of Hepatic Stem Cell Marker EpCAM by Wnt Signaling in Hepatocellular Carcinoma. Cancer Res 67(22):10831–10839. doi:10.1158/0008-5472 Imrich S, Hachmeister M, Gires O (2012) EpCAM and its potential role in tumor-initiating cells. Cell Adh Migr 6(1):30–38. doi:10.4161/cam.18953 Maetzel D, Denzel S, Mack B et al (2009) Nuclear signalling by tumour-associated antigen EpCAM. Nat Cell Biol 11(2): 162–171. doi:10.1038/ncb1824 Went PT, Lugli A, Meier S et al (2004) Frequent EpCam protein expression in human carcinomas. Hum Pathol 35(1):122–128. doi:10.1016/j.humpath.2003.08.026 Subira´ D, Serrano C, Castan˜o´n S et al (2012) Role of flow cytometry immunophenotyping in the diagnosis of leptomeningeal carcinomatosis. Neuro Oncol 14(1):43–52. doi:10.1093/ neuonc/nor172 Kurtz JE, Dufour P (2010) Adecatumumab: an anti-EpCAM monoclonal antibody, from the bench to the bedside. Expert Opin Biol Ther 10(6):951–958. doi:10.1517/14712598.2010.482098 Chaudhry IH, O’Donovan DG, Brenchley PE, Reid H, Roberts IS (2001) Vascular endothelial growth factor expression correlates with tumour grade and vascularity in gliomas. Histopathology 39:409–415. doi:10.1046/j.1365-2559.2001.01230.x Rahmah NN, Sakai K, Sano K, Hongo K (2012) Expression of RECK in endothelial cells of glioma: comparison with CD34 and VEGF expressions. J Neurooncol 107(3):559–564. doi:10.1007/ s11060-011-0778-z Netto GC, Bleil CB, Hilbig A, Coutinho LM (2008) Immunohistochemical evaluation of the microvascular density through the expression of TGF-beta (CD 105/endoglin) and CD 34 receptors and expression of the vascular endothelial growth factor (VEGF) in oligodendrogliomas. Neuropathology 28:17–23. doi:10.1111/j.1440-1789.2007.00825.x Vermeulen PB, Gasparini G, Fox SB et al (1996) Quantification of angiogenesis in solid human tumours: an international consensus on the methodology and criteria of evaluation. Eur J Cancer 32A:2474–2484 Deb P, Boruah D, Dutta V (2012) Morphometric study of microvessels in primary CNS tumors and its correlation with tumor types and grade. Microvasc Res 84(1):34–43. doi:10.1016/ j.mvr.2012.03.004 Varga Z, Diebold J, Dommann-Scherrer C et al (2012) How reliable is Ki-67 immunohistochemistry in grade 2 breast carcinomas? A QA study of the Swiss Working Group of Breast-and Gynecopathologists. PLoS One 7:e37379. doi:10.1371/journal. pone.0037379
J Neurooncol 32. Scholzen T, Gerdes J (2000) The Ki-67 protein: from the known and the unknown. J Cell Physiol 182:311–322. doi:10.1002/ (SICI)1097-4652(200003)182:3\311:AID-JCP1[3.0.CO;2-9 33. Hou XZ, Liu W, Fan HT et al (2010) Expression of hepatocyte growth factor and its receptor c-Met in human pituitary adenomas. Neuro Oncol 12(8):799–803. doi:10.1093/neuonc/noq024 34. Pang B, Fan H, Zhang IY et al (2012) HMGA1 expression in human gliomas and its correlation with tumor proliferation, invasion and angiogenesis. J Neurooncol 106(3):543–549. doi:10. 1007/s11060-011-0710-6 35. Fong D, Steurer M, Obrist P et al (2008) Ep-CAM expression in pancreatic and ampullary carcinomas: frequency and prognostic relevance. J Clin Pathol 61(1):31–35. doi:10.1136/jcp.2006.037333
36. Varga Meri, Obrist Peter, Schneeberger Stephan et al (2004) Overexpression of epithelial cell adhesion molecule antigen in gallbladder carcinoma is an independent marker for poor survival. Clin Cancer Res 10:3131–3136. doi:10.1158/1078-0432 37. Spizzo G, Went P, Dirnhofer S, Obrist P et al (2006) Overexpression of epithelial cell adhesion molecule (Ep-CAM) is an independent prognostic marker for reduced survival of patients with epithelial ovarian cancer. Gynecol Oncol 103(2):483–488. doi:10.1016/j.ygyno.2006.03.035 38. McGirt MJ, Chaichana KL, Gathinji M et al (2009) Independent association of extent of resection with survival in patients with malignant brain astrocytoma. J Neurosurg 110(1):156–162. doi:10.3171/2008.4.17536
123
