Accepted Manuscript

The EpCAM overexpression is associated with clinicopathological significance and

prognosis in hepatocellular carcinoma patients: A systematic review and meta-
analysis

Linhua Zhou, Yanping Zhu

PII: S1743-9191(18)31518-8
DOI: 10.1016/j.ijsu.2018.06.025
Reference: IJSU 4703

To appear in: International Journal of Surgery

Received Date: 9 March 2018

Revised Date: 7 June 2018
Accepted Date: 10 June 2018

Please cite this article as: Zhou L, Zhu Y, The EpCAM overexpression is associated with
clinicopathological significance and prognosis in hepatocellular carcinoma patients: A systematic review
and meta-analysis, International Journal of Surgery (2018), doi: 10.1016/j.ijsu.2018.06.025.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.
 ACCEPTED MANUSCRIPT

Title page
The EpCAM overexpression is associated with the clinicopathological
significance and prognosis in hepatocellular carcinoma patients: A systemic
review and meta-analysis

Linhua Zhou1, Yanping Zhu2*

PT
1
First Affiliated Hospital of Guangxi Medical University, Nanning, Guangxi Zhuang
Autonomous Region 530021, P. R. China

RI
2
Department of gastroenterology, Qiqihar first hospital, Qiqihar, Heilongjiang
Province, 161005, P.R. China

SC
*Correspondence: Yanping Zhu, Email: zhuyanping1773@163.com

U
AN
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT
The EpCAM overexpression is associated with clinicopathological significance
and prognosis in hepatocellular carcinoma patients: A systematic review and
meta-analysis
Abstract
Background
Hepatocellular carcinoma (HCC) is one of most common causes for cancer-related death around
the world. Epithelial cell adhesion molecule(EpCAM) is established as a vital prognostic factor for
the human malignant tumors. However, the potential role of EpCAM in HCC has largely remained

PT
elusive. Herein we aim to gain insight into the clinicopathological and prognostic role of EpCAM
in HCC.
Method and materials

RI
The PubMed, Web of Science, EMBASE and SCOPUS databases were systematically searched
from their inception up to 5 December 4, 2017. The hazard ratio(HR) and odds ratio(OR) were
respectively used as the effect size to explore the associations between EpCAM expression and the

SC
prognosis and clinicopathological features in HCC patients.
Results
Sixteen studies recruiting 2488 HCC patients were included in the meta-analysis, of which the

U
publication year ranging from 2011 to 2017. As a result, the pooled HR of 1.634 indicated that
AN
higher EpCAM expression was significantly associated with the shorter overall survival(OS)
periods (95%CIs: 1.151~2.320; Z=2.740, P=0.006). Next, a meta-analysis of disease-free
survival(DFS) was performed for the ten studies. Consequently, for the p-value less than 0.05 for
the combined HR, the overexpression of EpCAM was significantly correlated with poorer DFS.
M

Next, the results derived from our study suggest that the overexpression of EpCAM is associated
with the clinicopathological features of HCC, including poorer tumor differentiation and high
D

alpha-fetoprotein(AFP) levels.
Conclusion
TE

The results derived from our study suggest that the overexpression of EpCAM is associated with
the clinicopathological features of HCC, including poorer differentiation and high AFP levels.
More importantly, overexpression of EpCAM was confirmed as the unfavorable predictor for the
EP

shorter OS and DFS for HCC patients.

Keywords: EpCAM; Hepatocellular carcinoma; Prognosis; Clinicopathological features;
Meta-analysis
C
AC

1
 ACCEPTED MANUSCRIPT
1. Introduction
Hepatocellular carcinoma (HCC) is one of the most frequently diagnosed cancers worldwide.
There are overall 782500 new HCC cases in 2016, in which the number of males is twice more
than that of females[1]. This is mainly due to the prevalence of hepatitis B (HBV) and C (HCV)
viruses, together with alcoholic and nonalcoholic steatohepatitis[2-4]. Besides, HCC is a highly
fatal cancer that ranked fifth of cancer death in the United States in 2016. People dead of HCC
obviously increased at the highest rate of all types of cancer. These imply the growing burden of
liver cancer resulting in a heavy load on society[5]. Nevertheless, the early diagnosis of HCC is

PT
still difficult because of the absence of symptoms in the early stage of this disease. Therefore,
most of the patients are diagnosed in advanced-stage HCCs, leading to treatment limitations of
these advanced-stage patients. The early-stage HCC patients are mainly treated by surgery, while

RI
advanced HCC patients with no opportunity of resection have a limitation of systemic therapies,
such as sorafenib, who have a poor prognosis[6]. To improve the survival outcome of HCC
patients, it is critical to find novel biomarkers that can predict the survival outcome and provide a

SC
powerful tool for clinical HCC patients to choose the right treatment and avoid unnecessary
adverse reactions of inappropriate treatment.
Epithelial cell adhesion molecule (EpCAM), a membrane glycoprotein, is expressed in most

U
normal epithelial cells including fetal lung, liver, skin, and germ cells, and in adult epithelia[7] .
AN
EpCAM was proved to be involved in biological progress like cell proliferation, migration, and
mitogenic signal transduction[8]. It is also suggested that EpCAM is not only involved in cell
adhesion, but also cellular signaling. EpCAM is frequently expressed in human hepatic stem cells
in the liver but in mature hepatocytes[9]. Notably, EpCAM was found to be abnormal in many
M

epithelial cell cancers including HCC[10]. Recent studies also demonstrated that higher EpCAM
levels predicted the poor survival outcome in ovarian[11], colon[8], squamous head and neck
D

carcinoma[12] as well as breast cancers[13]. Previous reports showed that EpCAM overexpression
is associated with the angiogenesis and unfavorable outcome of HCC. Yamashita and his
TE

colleague revealed that EpCAM-positive cells were defined as valuable prognostic subtypes of
human HCC[14]. However, the function of EpCAM and its clinical significance of in HCC
remains largely unknown.
EP

In this study, we analyzed the clinicopathological significance of EpCAM in HCC and

correlated its expression with the prognosis. We attempted to perform a meta-analysis to evaluate
its role as a prognostic factor for survival in HCC entity.
C

2. Methods and materials

2.1 Search strategy and inclusion criteria
AC

The PubMed, Web of Science, EMBASE and SCOPUS databases were systematically
searched from their inception up to December 4, 2017. The search terms used for the databases
was shown as following: (liver neoplasms OR "Neoplasms, Liver" OR Liver Neoplasm OR
Hepatic Neoplasms OR Hepatic Neoplasm OR Cancer of Liver OR Hepatocellular Cancer OR
Hepatocellular Cancers OR Hepatic Cancer OR "Cancer, Hepatic" OR Hepatic Cancers OR Liver
Cancer OR "Cancer, Liver" OR Liver Cancers OR Cancer of the Liver) AND ("Epithelial Cell
Adhesion Molecule" OR "EPCAM "). The published article considered as eligibility must meet
the following inclusion criteria: (1) studies enrolled the patients diagnosed with HCC by clinical
pathologists; (2) the prospective or retrospective studies explored the association of EpCAM
expression and any of the following types of survival analysis in HCC patients: overall

2
 ACCEPTED MANUSCRIPT
survival(OS) and disease-free survival (DFS, also called recurrence-free, RFS); (3) the expression
of EpCAM levels were detected by quantitative real time –PCR, immunohistochemistry,
microarray or ELISA; (4) the studies provided the information associated to the survival analysis,
to estimate the hazard risk(HR) and 95% confidence intervals(CIs). The following studies were
considered as ineligible: reviews, meeting abstracts, letters, experimental studies, and studies
without sufficient information. The flowchart of the study selection was illustrated in the
Figure1.
2.2 Data extraction and quality assessment

PT
Two authors carefully evaluated the title and abstracts and further reviewed the full-text. The
key information from the studies were extracted into baseline tables: first author, and published
year, publication country, patients number, cutoff value, treatment, specimen type, EpCAM

RI
detection assay, HR estimated method and HRs and 95% CIs. The Newcastle-Ottawa Scale (NOS)
was used to assess the quality of the included studies[15]. All the processes were in accordance
with the PRISMA (Preferred Reporting Items for Systemic Reviews and Meta-analysis)

SC
guidelines[16] and Cochrane handbook for systematic reviews of interventions[17].
2.3 Statistical methods
Summarized HRs and 95% CIs of EpCAM overexpression were used to evaluate the

U
predicted ability for the prognosis of HCC patients. As previously described [27], three types of
AN
the effect sizes were derived from the original study: (1) multivariable HRs adjusting for other
prognostic factor; (2) univariable HRs; (3) HRs calculated from Kaplan–Meier(KM) curves. As
for KM curves, the total event, the log-Rank P value and the primary patients were collected to
calculate the HRs as the previous studies described[18, 19]. Importantly, in the study providing the
M

multivariable HR and univariable HR for the same cohort, the multivariable HRs were obtained to
better reveal the independent effort of EpCAM for HCC patients. The pooled HRs for OS and RFS
D

were estimated separately. A combined HR larger than 1 without its 95%CIs overlapping 1
suggested the unfavorable outcome for the HCC patients with high EpCAM levels. The statistic
TE

model for combining HRs was dependent on the study variation (heterogeneity), in which the
random effect model was performed when heterogeneity was significant (P-value <0.1 or I2>50%).
The fixed effect model was conducted when the P-value of heterogeneity was greater than 0.1
EP

across the studies[20]. Further, the robustness of the meta-analysis was evaluated by the sensitivity
analysis[21]. Finally, the publication bias across the studies was also assessed by the funnel plot
and Begg’ test[22]. A statistical two-tailed P value < 0.05 was defined as a statistical significance.
C

3. Results
3.1 Study characteristics
AC

As a result, a total of 1969 records were initially identified in the four databases according to
the search strategy. Next, 841 duplicated studies were excluded, and 1155 studies remained for the
further evaluation. After screening the titles and abstracts, 60 studies considered as potentially
eligible were further reviewed in full-text. Finally, sixteen studies[23-38] recruiting 2488 HCC
patients were included in the meta-analysis, of which the publication year ranging from 2011 to
2017. In details, fourteen studies evaluated the OS information and ten studies reported the DFS
data for HCC patients. Nine of the sixteen studies were carried out from China, and two were from
Korea. As shown in the Table1, most of the studies enrolled the patients after the hepatectomy
treatment. Of note, three studies detected the EpCAM levels in the blood sample while the
remaining studies utilized the tissues samples. In the studied performing the

3
 ACCEPTED MANUSCRIPT
immunohistochemistry, the cut-off value for dividing EpCAM expression into high levels and low
levels were ranging from 1% to 25%.
3.2 The prognostic value of EpCAM for HCC patients.
In the 14 articles reporting on OS, a total of 2338 HCC patients were enrolled for the analysis.
As a result, the pooled HR of 1.634 indicated that higher EpCAM expression was significantly
associated with the shorter OS periods (95%CIs: 1.151~2.320; Z=2.740, P=0.006). The obvious
heterogeneity was found across these 14 studies so that the random effect model was used to
combine the HRs (I2=83.30%, P<0.005，Figure2A). Next, a meta-analysis of DFS was performed

PT
for the ten studies. Consequently, for the combined HR of 1.558, the overexpression of EpCAM
was significantly correlated with poorer DFS(Figure2B). There was no significant heterogeneity in
the analysis and the fix effect model was utilized. Next, we also aimed to explore the independent

RI
prognostic implications of EpCAM in the HCC patients by conducting the subgroup analysis.
Notably, the combined effects for multivariate HRs on OS and DFS, indicated the EpCAM acts as
an independent prognostic factor for HCC patients (HR=2.327 for OS, P<0.001 and HR=1.604,

SC
P<0.005). Additionally, no significant heterogeneity was found in these two analyses.
3.3 The correlations between EpCAM overexpression and clinicopathological parameters in
HCC patients

U
Subsequently, the associations between EpCAM overexpression and the clinicopathological
AN
features of HCC patients were then investigated. A summary OR of 2.359 (95%CIs: 1.482~
3.754; Z=3.62, P<0.005) studies were identified in the 4 studies reporting the differentiation,
suggesting the positive correlation between high EpCAM levels and the presence of poor
differentiation (OR=2.36, 95%CIs: 1.48~3.75; I2=3.4%, P=0.375). More importantly, among the
M

studies reporting AFP levels, seven studies with 952 patients suggested an increase of AFP level in
the patients with EpCAM positive compared with the negative ones, suggesting the potential role
D

of EpCAM for the diagnosis of HCC patients (OR=2.252, 95%CIs: 1.641~3.092, Z=5.03, p
=0.000; I2=0.0%, P= 0.499, Figure3).
TE

3.4 Sensitivity analysis

The sensitivity analysis was respectively performed for the OS and DFS studies. As a result,
the result derived from the analysis was robust.
EP

3.5 Publication bias

As illustrated in the Supplementary Figure1, there was no obvious publication bias.
4. Discussion
C

The main finding of this study consists of two objectives. One is that EpCAM overexpression
is associated with the reduction in the survival of the HCC patients. Second, the effect of level
AC

EpCAM on the progression in HCC was also addressed, potentially leading to the poor
differentiation and higher AFP levels.
These findings are consistent with the previous studies that EpCAM overexpression is
correlated with the aggressiveness and progression of cancer patients. For instance, higher
EpCAM levels also represented the prognostic implications in predicting the unfavorable survival
outcome in oral squamous cell carcinoma[39]. The results investigating in lung cancer were also
in accordance with our study that EpCAM positive correlated s significantly with poor survival.
On the contrary, a more favorable outcome were found in the ovarian cancer patients with
EpCAM overexpression[40]. Recently, a meta-analysis carried out by Dai et al. revealed that the
patients with EpCAM overexpression increased the risk of tumor extension and lymph node

4
 ACCEPTED MANUSCRIPT
metastasis as well as worse prognosis in gastric cancer[41]. The meta-analysis conducted by Rui
Liu demonstrated that the positive expression of EpCAM was associated with poor differentiation
of HCC while the included studies were comparatively few[42]. In our studies, we included four
studies with 604 patients for the investigating the correlation between poor differentiation and
EpCAM. Previously, Ma and his group has published a relevant meta-analysis that explored the
potential prognostic value of stem cell markers for HCC, including CD133 and EpCAM. However,
the result showed that EpCAM could not reach a statistical significance for its correlation and the
survival of HCC patients[43]. The main reason for discrepancy is that this previous study only

PT
collected the data from four studies, which was unpersuasive. So far as we acknowledged, our
study enrolled the largest number of included studies(n=16) and sample size(n=2488) to explore
the potential effect of EpCAM on the clinicopathological features and prognosis of HCC. In the

RI
support of the previous studies, we strongly supported that EpCAM acts as a vital biomarker and a
valuable prognostic factor for HCC. Consistent with our findings, Xu Minhui and his colleagues
reported that HCC patients with high EpCAM expression showed higher relapse rates and were

SC
more likely to be associated with aggressive clinical features, such as high AFP levels, advanced
TNM stages[44]. It is known that EpCAM is involved in maintenance of hepatic cancer stem cell
by interacting with Wnt–β-catenin signaling[45]. EpCAM-regulated intramembrane proteolysis

U
and Wnt pathway activation mediate the expression of the hCSC-like (hepatic cancer stem cells)
AN
gene signature [46]. The PTEN/Akt/mTOR pathway might play an essential role in driving
recurrence and influencing prognosis in HCC. Moreover, the combination of PTEN with CD133
or EpCAM expression may serve as a screening tool to monitor recurrence and predict
prognosis[47]. It is reported that EpCAM is negatively up-regulated by the wild-type p53, which
M

is a well-established tumor suppressor in tumor[48]. A previous study revealed that higher

EpCAM expression levels contributed to chemoresistance in HCC[49]. It has also been suggested
D

that the intracellular domain of EpCAM, as a transcriptional factor, induces cell activation
molecules like c-myc, promoting cell proliferation[50]. Bea and his colleagues silenced EpCAM
TE

by siRNA in Huh-1 cells, resulting in the inhibition of spheroid formation and tumorigenicity[38].
Nevertheless, the underlying mechanisms how EpCAM is involved in the progression and affect
the survival outcome are largely unknown.
EP

However, certain limitations in this study should be taken into account. First, the detection
assay is very important. Most of the included studies detected the EpCAM expression using IHC,
which is widely used for the protein assessment. While there were several studies performed the
C

other technique like qRT-PCR, microarray as well as the semiauto-mated fluorescence-based

microscopy, potentially increasing the heterogeneity. Second, the varied cutoff values for EpCAM
AC

expression measured by the IHC method determined by the author could also lead to the
discrepancy across the studies, which is lack of a standard determination. Three articles used 5%
of total tumor cells as the cutoff value when using the IHC method, however, the other studies
made 1%, 2%, and 25% as the cutoff value. Third, the different types of the specimens used in the
original studies might also be a potential bias for the assessment of EpCAM level though most of
the studies used the tissues samples. Fourth, six studies on OS and two studies on DFS did not
provide neither estimated univariate HRs nor multivariate HRs so we could only calculate the
effect size from the KM curves, resulting the inaccuracy for the data collection. Fifth, one of the
inclusion criteria on the language restriction would also leave out the eligible studies that were
written in the other languages for that the negative results were preferred to be published in the

5
 ACCEPTED MANUSCRIPT
local journals.
Taken together, the results derived from our study suggested that the overexpression of
EpCAM is associated with the clinicopathological features of HCC, including poor differentiation
as well as high AFP levels. More importantly, overexpression of EpCAM was confirmed as the
unfavorable predictor for the OS and DFS for HCC patients. However, the further experimental
study and clinical research are needed to pave the way for uncovering the potential biological
mechanisms of EpCAM for HCC.

PT
Abbreviation
Hepatocellular carcinoma (HCC); hepatitis B viruses (HBV); Epithelial cell adhesion molecule
(EpCAM); Hazard risk(HR); overall survival(OS); Disease-free survival(DFS); Confidence

RI
intervals(CIs); Kaplan–Meier(KM).

SC
Funding
This research did not receive any specific grant from funding agencies in the public, commercial,
or not-for-profit sectors.

U
Acknowledgments
AN
None

Conflicts of interest
The authors declare that they have no conflict of interest.
M
D
TE
C EP
AC

6
 ACCEPTED MANUSCRIPT
References
1. Siegel RL, Miller KD, Jemal A: Cancer Statistics, 2017. CA Cancer J Clin 2017, 67(1):7-30.
2. Perz JF, Armstrong GL, Farrington LA, Hutin YJ, Bell BP: The contributions of hepatitis B
virus and hepatitis C virus infections to cirrhosis and primary liver cancer worldwide.
Journal of hepatology 2006, 45(4):529-538.
3. Chuang S-C, La Vecchia C, Boffetta P: Liver cancer: descriptive epidemiology and risk
factors other than HBV and HCV infection. Cancer letters 2009, 286(1):9-14.
4. Smedile A, Bugianesi E: Steatosis and hepatocellular carcinoma risk. European review for

PT
medical and pharmacological sciences 2005, 9(5):291.
5. Ryerson AB, Eheman CR, Altekruse SF, Ward JW, Jemal A, Sherman RL, Henley SJ,
Holtzman D, Lake A, Noone AM: Annual report to the nation on the status of cancer,

RI
1975 ‐ 2012, featuring the increasing incidence of liver cancer. Cancer 2016,
122(9):1312-1337.

SC
6. Bruix J, Gores GJ, Mazzaferro V: Hepatocellular carcinoma: clinical frontiers and
perspectives. Gut 2014:gutjnl-2013-306627.
7. Litvinov SV, Velders MP, Bakker H, Fleuren GJ, Warnaar SO: Ep-CAM: a human epithelial
antigen is a homophilic cell-cell adhesion molecule. The Journal of cell biology 1994,

U
125(2):437-446.
AN
8. Munz M, Baeuerle PA, Gires O: The emerging role of EpCAM in cancer and stem cell
signaling. Cancer research 2009, 69(14):5627-5629.
9. Winter MJ, Nagtegaal ID, van Krieken JHJ, Litvinov SV: The epithelial cell adhesion
molecule (Ep-CAM) as a morphoregulatory molecule is a tool in surgical pathology. The
M

American journal of pathology 2003, 163(6):2139-2148.

10. Litvinov SV, Balzar M, Winter MJ, Bakker HA, Briaire-de Bruijn IH, Prins F, Fleuren GJ,
D

Warnaar SO: Epithelial cell adhesion molecule (Ep-CAM) modulates cell–cell interactions
mediated by classic cadherins. The Journal of cell biology 1997, 139(5):1337-1348.
TE

11. Tas F, Karabulut S, Serilmez M, Ciftci R, Duranyildiz D: Clinical significance of serum

epithelial cell adhesion molecule (EPCAM) and vascular cell adhesion molecule-1
(VCAM-1) levels in patients with epithelial ovarian cancer. Tumor Biology 2014,
EP

35(4):3095-3102.
12. Pauli C, Münz M, Kieu C, Mack B, Breinl P, Wollenberg B, Lang S, Zeidler R, Gires O:
Tumor-specific glycosylation of the carcinoma-associated epithelial cell adhesion
C

molecule EpCAM in head and neck carcinomas. Cancer letters 2003, 193(1):25-32.
13. Osta WA, Chen Y, Mikhitarian K, Mitas M, Salem M, Hannun YA, Cole DJ, Gillanders WE:
AC

EpCAM is overexpressed in breast cancer and is a potential target for breast cancer gene
therapy. Cancer research 2004, 64(16):5818-5824.
14. Ji J, Yamashita T, Budhu A, Forgues M, Jia HL, Li C, Deng C, Wauthier E, Reid LM, Ye QH:
Identification of microRNA‐
‐181 by genome‐
‐wide screening as a critical player in
EpCAM–
–positive hepatic cancer stem cells. Hepatology 2009, 50(2):472-480.
15. Stang A: Critical evaluation of the Newcastle-Ottawa scale for the assessment of the
quality of nonrandomized studies in meta-analyses. European journal of epidemiology
2010, 25(9):603-605.
16. Moher D, Liberati A, Tetzlaff J, Altman DG, Group P: Preferred reporting items for
systematic reviews and meta-analyses: the PRISMA statement. PLoS medicine 2009,

7
 ACCEPTED MANUSCRIPT
6(7):e1000097.
17. Higgins JP, Green S: Cochrane handbook for systematic reviews of interventions. In.:
version; 2005.
18. Parmar MK, Torri V, Stewart L: Extracting summary statistics to perform meta‐
‐analyses
of the published literature for survival endpoints. Statistics in medicine 1998,
17(24):2815-2834.
19. Tierney JF, Stewart LA, Ghersi D, Burdett S, Sydes MR: Practical methods for
incorporating summary time-to-event data into meta-analysis. Trials 2007, 8(1):16.

PT
20. DerSimonian R, Laird N: Meta-analysis in clinical trials. Controlled clinical trials 1986,
7(3):177-188.
21. Sutton AJ, Abrams KR, Jones DR, Jones DR, Sheldon TA, Song F: Methods for

RI
meta-analysis in medical research. 2000.
22. Kicinski M, Springate DA, Kontopantelis E: Publication bias in meta‐
‐analyses from the

SC
Cochrane Database of Systematic Reviews. Statistics in medicine 2015, 34(20):2781-2793.
23. Ziol M, Sutton A, Calderaro J, Barget N, Aout M, Leroy V, Blanc J-F, Sturm N, Bioulac-Sage
P, Nahon P: ESM-1 expression in stromal cells is predictive of recurrence after
radiofrequency ablation in early hepatocellular carcinoma. Journal of hepatology 2013,

U
59(6):1264-1270.
AN
24. Xu M, Qian G, Xie F, Shi C, Yan L, Yu L, Zheng T, Wei L, Yang J: Expression of epithelial
cell adhesion molecule associated with elevated ductular reactions in hepatocellar
carcinoma. Clinics and research in hepatology and gastroenterology 2014, 38(6):699-705.
25. van Malenstein H, Komuta M, Verslype C, Vandecaveye V, Van Calster B, Topal B, Laleman
M

W, Cassiman D, Van Steenbergen W, Aerts R: Histology obtained by needle biopsy gives

additional information on the prognosis of hepatocellular carcinoma. Hepatology
D

Research 2012, 42(10):990-998.

26. Sun YF, Xu Y, Yang XR, Guo W, Zhang X, Qiu SJ, Shi RY, Hu B, Zhou J, Fan J: Circulating
TE

stem cell–like epithelial cell adhesion molecule–positive tumor cells indicate poor
prognosis of hepatocellular carcinoma after curative resection. Hepatology 2013,
57(4):1458-1468.
EP

27. Su R, Nan H, Guo H, Ruan Z, Jiang L, Song Y, Nan K: Associations of components of

PTEN/AKT/mTOR pathway with cancer stem cell markers and prognostic value of these
biomarkers in hepatocellular carcinoma. Hepatology Research 2016, 46(13):1380-1391.
C

28. Shibahara J, Ando S, Hayashi A, Sakamoto Y, Hesegawa K, Kokudo N, Fukayama M:

Clinicopathologic characteristics of SALL4-immunopositive hepatocellular carcinoma.
AC

Springerplus 2014, 3(1):721.

29. Schulze K, Gasch C, Staufer K, Nashan B, Lohse AW, Pantel K, Riethdorf S, Wege H:
Presence of EpCAM‐
‐positive circulating tumor cells as biomarker for systemic disease
strongly correlates to survival in patients with hepatocellular carcinoma. International
journal of cancer 2013, 133(9):2165-2171.
30. Murakata A, Tanaka S, Mogushi K, Yasen M, Noguchi N, Irie T, Kudo A, Nakamura N,
Tanaka H, Arii S: Gene expression signature of the gross morphology in hepatocellular
carcinoma. Annals of surgery 2011, 253(1):94-100.
31. Miltiadous O, Sia D, Hoshida Y, Fiel MI, Harrington AN, Thung SN, Tan PS, Dong H, Revill
K, Chang CY: Progenitor cell markers predict outcome of patients with hepatocellular

8
 ACCEPTED MANUSCRIPT
carcinoma beyond Milan criteria undergoing liver transplantation. Journal of hepatology
2015, 63(6):1368-1377.
32. Liang J, Ding T, Guo Z, Yu X, Hu Y, Zheng L, Xu J: Expression pattern of
tumour-associated antigens in hepatocellular carcinoma: association with immune
infiltration and disease progression. British journal of cancer 2013, 109(4):1031-1039.
33. Li L, Liu Y, Guo Y, Liu B, Zhao Y, Li P, Song F, Zheng H, Yu J, Song T: Regulatory MiR‐
‐
148a ‐ ACVR1/BMP circuit defines a cancer stem cell‐
‐ like aggressive subtype of
hepatocellular carcinoma. Hepatology 2015, 61(2):574-584.

PT
34. Guo Z, Li L-Q, Jiang J-H, Ou C, Zeng L-X, Xiang B-D: Cancer stem cell markers correlate
with early recurrence and survival in hepatocellular carcinoma. World journal of
gastroenterology: WJG 2014, 20(8):2098.

RI
35. Dai X-M, Huang T, Yang S-L, Zheng X-M, Chen GG, Zhang T: Peritumoral EpCAM Is an
Independent Prognostic Marker after Curative Resection of HBV-Related

SC
Hepatocellular Carcinoma. Disease markers 2017, 2017.
36. Choi GH, Kim GI, Yoo JE, Na DC, Han DH, Roh YH, Park YN, Choi JS: Increased
expression of circulating cancer stem cell markers during the perioperative period
predicts early recurrence after curative resection of hepatocellular carcinoma. Annals of

U
surgical oncology 2015, 22(3):1444-1452.
AN
37. Chan AW, Tong JH, Chan SL, Lai P, To KF: Expression of stemness markers (CD133 and
EpCAM) in prognostication of hepatocellular carcinoma. Histopathology 2014,
64(7):935-950.
38. Bae JS, Noh SJ, Jang KY, Park HS, Chung MJ, Park CK, Moon WS: Expression and role of
M

epithelial cell adhesion molecule in dysplastic nodule and hepatocellular carcinoma.

International journal of oncology 2012, 41(6):2150-2158.
D

39. Sen S, Carnelio S: Expression of epithelial cell adhesion molecule (EpCAM) in oral
squamous cell carcinoma. Histopathology 2016, 68(6):897-904.
TE

40. Woopen H, Pietzner K, Richter R, Fotopoulou C, Joens T, Braicu EI, Mellstedt H, Mahner S,
Lindhofer H, Darb-Esfahani S: Overexpression of the epithelial cell adhesion molecule is
associated with a more favorable prognosis and response to platinum-based
EP

chemotherapy in ovarian cancer. Journal of gynecologic oncology 2014, 25(3):221-228.

41. Dai M, Yuan F, Fu C, Shen G, Hu S, Shen G: Relationship between epithelial cell adhesion
molecule (EpCAM) overexpression and gastric cancer patients: A systematic review and
C

meta-analysis. PloS one 2017, 12(4):e0175357.

42. Liu R, Shen Y, Nan K, Mi B, Wu T, Guo J, Li M, Lv Y, Guo H: Association between
AC

expression of cancer stem cell markers and poor differentiation of hepatocellular

carcinoma: a meta-analysis (PRISMA). Medicine 2015, 94(31).
43. Ma Y-C, Yang J-Y, Yan L-N: Relevant markers of cancer stem cells indicate a poor
prognosis in hepatocellular carcinoma patients: a meta-analysis. European journal of
gastroenterology & hepatology 2013, 25(9):1007-1016.
44. Xu M, Qian G, Xie F, Shi C, Yan L, Yu L, Zheng T, Wei L, Yang J: Expression of epithelial
cell adhesion molecule associated with elevated ductular reactions in hepatocellar
carcinoma. Clin Res Hepatol Gastroenterol 2014, 38(6):699-705.
45. Yang W, Yan H-X, Chen L, Liu Q, He Y-Q, Yu L-X, Zhang S-H, Huang D-D, Tang L, Kong
X-N: Wnt/β-catenin signaling contributes to activation of normal and tumorigenic liver

9
 ACCEPTED MANUSCRIPT
progenitor cells. Cancer research 2008, 68(11):4287-4295.
46. Mani SK, Zhang H, Diab A, Pascuzzi PE, Lefrancois L, Fares N, Bancel B, Merle P, Andrisani
O: EpCAM-regulated intramembrane proteolysis induces a cancer stem cell-like gene
signature in hepatitis B virus-infected hepatocytes. J Hepatol 2016, 65(5):888-898.
47. Su R, Nan H, Guo H, Ruan Z, Jiang L, Song Y, Nan K: Associations of components of
PTEN/AKT/mTOR pathway with cancer stem cell markers and prognostic value of these
biomarkers in hepatocellular carcinoma. Hepatology research : the official journal of the
Japan Society of Hepatology 2016, 46(13):1380-1391.

PT
48. Sankpal NV, Willman MW, Fleming TP, Mayfield JD, Gillanders WE: Transcriptional
repression of epithelial cell adhesion molecule contributes to p53 control of breast cancer
invasion. Cancer research 2009, 69(3):753-757.

RI
49. Kimura O, Kondo Y, Kogure T, Kakazu E, Ninomiya M, Iwata T, Morosawa T, Shimosegawa
T: Expression of EpCAM increases in the hepatitis B related and the treatment-resistant

SC
hepatocellular carcinoma. BioMed research international 2014, 2014.
50. Maetzel D, Denzel S, Mack B, Canis M, Went P, Benk M, Kieu C, Papior P, Baeuerle PA,
Munz M: Nuclear signalling by tumour-associated antigen EpCAM. Nature cell biology
2009, 11(2).

U
AN
M
D
TE
C EP
AC

10
 ACCEPTED MANUSCRIPT

Table1 Characteristics of included studies

PT
Author/year Year Region Histological Patients Cutoff value Treatment Sample Assay NOS score
type

RI
Marianne Ziol 2013 France HCC 150 0.05 Radiofrequency Tissue Immunohistochemistry 9
ablation

SC
Hannah van 2012 Belgium HCC 58 0.05 RFA or TACE Tissue Immunohistochemistry 7
Malenstein
Rujuan Su 2016 China HCC 110 0.25 Hepatectomy Tissue Immunohistochemistry 9

U
Zhe Guo 2016 China HCC 50 0.05 Hepatectomy Tissue Immunohistochemistry 6

AN
Yun-Fan Sun 2013 China HCC 123 2 Hepatectomy Blood qRT-PCR 9
Minhui Xu 2014 China HCC 106 NA Hepatectomy Tissue Immunohistochemistry 9
J Liang 2013 China HCC 362 NA Hepatectomy Tissue microarray 6

M
Ayano Murakata 2011 China HCC 275 NA Hepatectomy Tissue Immunohistochemistry 6
XiaoMeng Dai 2017 China HCC 106 NA Hepatectomy Tissue microarray 9

D
Lian Li 2015 China HCC 100 NA Hepatectomy Tissue Immunohistochemistry 7

TE
Kornelius Schulze 2013 Germany HCC 59 1 CTC/7.5 NA Blood semiauto-mated 6
mL fluorescence-based
microscopy
EP
Anthony W H Chan 2014 Hong HCC 282 0.01 Hepatectomy Tissue Immunohistochemistry 9
Kong
C

Junji Shibahara 2014 Japan HCC 337 0.02 Hepatectomy Tissue Immunohistochemistry 8
AC

Jun Sang Bae 2012 Korea HCC 175 0.6 Hepatectomy Tissue Immunohistochemistry 9
Gi Hong Choi 2015 Korea HCC 63 NA Hepatectomy blood qRT-PCR 6
Oriana Miltiadous 2015 USA HCC 132 NA NA Tissue Immunohistochemistry 6

11
 ACCEPTED MANUSCRIPT

Table2 Prognostic value of EpCAM for OS and DFS in HCC patients.

Author/year Year HR LL UL Survival Extract method HR LL UL Survival

PT
Marianne Ziol 2013 Univariate 1.480 0.750 2.900 DFS
Hannah van Malenstein 2012 1.980 0.880 4.020 OS

RI
Rujuan Su 2016 1.368 0.782 3.653 OS Multivariate 2.149 1.218 5.308 DFS
Zhe Guo 2016 3.687 1.366 9.949 OS

SC
Yun-Fan Sun 2013 5.200 2.650 10.210 OS
Minhui Xu 2014 2.031 1.191 3.463 OS Multivariate 1.708 1.089 2.679 DFS
J Liang 2013 0.584 0.446 0.766 OS

U
Ayano Murakata 2011 2.284 1.238 4.213 OS KM 2.343 1.306 4.203 DFS

AN
XiaoMeng Dai 2017 2.030 1.252 3.290 OS Multivariate 1.622 0.986 2.669 DFS
Lian Li 2015 2.000 1.300 3.000 OS Univariate 1.200 0.700 2.000 DFS
Kornelius Schulze 2013 2.122 1.144 3.938 OS

M
Anthony W H Chan 2014 1.328 0.849 2.077 OS Multivariate 2.048 1.350 3.108 DFS
Junji Shibahara 2014 1.615 0.884 2.951 OS Multivariate 1.133 0.768 1.672 DFS

D
Jun Sang Bae 2012 1.022 0.679 1.537 OS

TE
Gi Hong Choi 2015 KM 2.302 0.692 7.658 DFS
Oriana Miltiadous 2015 0.497 0.210 1.150 OS Univariate 0.702 0.300 1.660 DFS
C EP
AC

12
 ACCEPTED MANUSCRIPT

Table3 Overall and subgroup analysis of EpCAM for OS and DFS in HCC cancer patients.
Variables HR(95%CIs) Z P I2 P Statistical model

PT
Overall survival 1.634(1.151~2.320) 2.740 0.006 83.30% 0 Random-effect
Extract method

RI
Univariate HR 1.431(0.835~2.451) 1.310 0.192 65.40% 0.034 Random-effect
Multivariate HR 2.327(1.430~3.785) 3.400 0.001 61.70% 0.05 Random-effect

SC
KM Curve 1.435(0.842~2.444) 1.330 0.184 86.10% 0 Random-effect
DFS 1.558(1.312~1.849) 5.06 0 23.30% 0.229 Random-effect
Extract method

U
Univariate HR 1.155(0.795~1.677) 0.76 0.449 0.00% 0.399 Fix-effect

AN
Multivariate HR 1.604(1.303~1.975) 4.46 0 21.10% 0.28 Fix-effect
KM Curve 2.335(1.381~3.950) 5.06 0 0.00% 0.979 Fix-effect

M
D
TE
C EP
AC

13
 ACCEPTED MANUSCRIPT
Figure Legends
Figure1 Flow chart of study identification process.
Figure2 Forest plot of the effect of EpCAM status on survival. A. The pooled HR of 1.634
indicated that higher EpCAM expression was significantly associated with the shorter OS periods
(95%CIs: 1.151~2.320; Z=2.740, P=0.006); B. The pooled HR of 1.558 indicated that the
overexpression of EpCAM was statistically significantly correlated with poorer DFS (95%CIs:
1.312~1.849; Z=5.06, P=0.000)
Figure3 Forest plot of the effect of EpCAM status on clinicopathological features. A. The

PT
positive correlation between high EpCAM levels and the present of poor differentiation (OR=2.36,
95%CIs:1.48~3.75; I23.4%, P=0.375); B. An increase of AFP level in the patients with EpCAM
positive compared with the negative ones, suggesting the potential role of EpCAM for the

RI
diagnosis of HCC patients (OR=2.252, 95%CIs: 1.641~3.092, Z=5.03 p =0.000; I2=0.0%, P=
0.499)
Supplementary Figure1 Funnel plot of the analysis on survival. A. There was no obvious

SC
publication bias for the outcome of OS; B. There was no obvious publication bias for the outcome
of DFS

U
AN
M
D
TE
C EP
AC

14
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT
Highlights:
* High EpCAM expression is an unfavorable factor for OS and DFS in HCC.
* Overexpression of EpCAM is associated with the clinicopathological features of HCC.

* EpCAM expression can predict biological behavior of HCC.

* EpCAM expression can predict prognosis of HCC.

PT
RI
U SC
AN
M
D
TE
C EP
AC