RESEARCH ARTICLE

Clinical significance of serum epithelial cell adhesion molecule(EPCAM) and vascular cell adhesion molecule-1 (VCAM-1)levels in patients with epithelial ovarian cancer

Faruk Tas & Senem Karabulut & Murat Serilmez &

Rumeysa Ciftci & Derya Duranyildiz

Received: 19 August 2013 /Accepted: 6 November 2013# International Society of Oncology and BioMarkers (ISOBM) 2013

Abstract Cellular adhesion molecules might be goodmarkers in some types of malignant tumors, useful informa-tion in diagnosis and prognosis. The objective of this studywas to evaluate the serum levels of epithelial cell adhesionmolecule (EPCAM) and vascular cell adhesion molecule-1(VCAM-1) in epithelial ovarian cancer (EOC) patients. Fiftypatients were enrolled into the study. Serum EPCAM andVCAM-1 levels were determined by the solid-phase sandwichELISA method. Age- and sex-matched 30 healthy controlswere included in the analysis. The median age of the patientswas 56.5 years, range 22 to 83 years. Majority of the patientshad advanced disease (stages III–IV) (90 %). The baselineserum EPCAM levels of the EOC patients were significantlyhigher than in those in the control group (p =0.03). However,there was no significant difference in the serum VCAM-1level between EOC patients and controls (p =0.24).Metastatic patients had higher serum VCAM-1 levels com-pared with the non-metastatic patients (p =0.03). Moreover,no other clinical variables including response to chemotherapywere found to be correlated with both serum assays (p >0.05).No correlation was found between serum EPCAM andVCAM-1 levels in EOC patients (r s=0.105, p =0.362).Neither serum EPCAM level nor serum VCAM-1 level hadsignificant adverse effect on survival. In conclusion, the higherbaseline serum levels of VCAM-1 were associated with met-astatic disease, and serum EPCAM level was found to be adiagnostic marker in EOC patients. However, both serumassays had no prognostic roles on outcome.

Keywords EPCAM .VCAM-1 . Serum . Ovarian cancer

Introduction

Epithelial ovarian cancer (EOC) is the fourth most frequentcause of death from cancer and the most fatal gynecologicmalignancy in women. Because EOC typically cause fewspecific symptoms, more than two thirds of patients are diag-nosed with advanced disease, 5-year survival rates are lessthan 30 %. Therefore, early detection of EOC has greatpromise to improve clinical outcome. Similar to serum CA125, several EOC-related serum protein biomarkers lack suf-ficient sensitivity and specificity for detection of early stagedisease. Thus, the need to develop a diagnostic assay in a largeEOC diagnosed population still exists.

Cellular adhesion molecules (CAMs) mediate a greatvariety of cellular interactions in normal tissue physiologyand process of carcinogenesis [1]. In normal, CAMs partici-pate in tissue growth and maintenance, blood coagulation,wound healing, and inflammation. Moreover, they are in-volved in invasion and metastasis in malign neoplastic trans-formation. In oncology practice, CAMs might be goodmarkers in some types of malignant tumors, useful informa-tion in diagnosis and prognosis [1].

Epithelial cell adhesionmolecule (EPCAM) is a transmem-brane, epithelial-specific intercellular cell adhesion molecule[2–11]. Although its function is still largely unknown, it isinvolved also in cellular signaling, cell migration, prolifera-tion, and differentiation in addition to cell adhesion. In human,EPCAM is expressed exclusively in epithelia and epithelial-derived neoplasms. Therefore, it can be used as a histologi-cally diagnostic and differentiative marker for epithelial tu-mors from non-epithelial tumors. EPCAM expression is notfound in mesenchymal, muscular, neuroendocrine, lymphoid,and melanoma-originated tumor. EPCAM is detected by im-munohistochemistry (IHC) in histological tumor samples [2].More specifically, EPCAM is detected at the basolateral cell

F. Tas (*) : S. Karabulut :M. Serilmez : R. Ciftci :D. DuranyildizInstitute of Oncology, University of Istanbul, Capa, 34390 Istanbul,Turkeye-mail: faruktas2002@yahoo.com

Tumor Biol.DOI 10.1007/s13277-013-1401-z

membrane of normal human tissues [2]. EPCAM is moreintensely positive in tumor cells than in the normal tissue.High EPCAM expression has been reported in various epithe-lial tumors such as, colorectal, pancreas, stomach, breast,prostate, lung cancer, and ovarian cancer [2]. In other typesof cancer including ovarian cancer, the role of EPCAM asdiagnostic is not clear, and contradictory results have beenreported (2–11). Although a high expression level of EPCAMhas been reported in ovarian cancer, the clinical implicationhas not been determined fully [2–11]. So far, the diagnostic,prognostic, and predictive marker roles of EPCAM are un-clear in ovarian cancer.

Vascular cell adhesion molecule (VCAM) is a member ofthe immunoglobulin gene superfamily, which is a cell surfacemolecule sharing a segmental structural region consisting of70–110 amino acids in the immunoglobulin chain [1].VCAM-1 is constitutively expressed on many different typesof endothelial and stromal cells, and mediates cellular adhe-sion. VCAM-1 appears to be of special interest because it maybe involved in the process of invasion and metastatic spread.Although high expression of VCAM-1 has been shown to beassociated with several cancers, and increased expression hasbeen found to be related to progression of several cancers, itsrole in ovarian cancer is largely undefined [12–14]. Similar toother tumors, the expression of VCAM-1 has been also dem-onstrated histologically in tumor cell lines of EOC patients[15, 16]. Elevated serum levels of VCAM-1 have been iden-tified in various malignant diseases, including breast cancer[17, 18] and colorectal cancer [19–21]. In these studies, thesoluble VCAM-1 levels in the patients were significantlyhigher than those in the control subjects. The increased solublelevels of VCAM-1 were associated with advanced stage andprognostic significance. However, as yet, the serum levels ofVCAM-1 have not been compared with tumor stage andtumor cell expression of VCAM-1in EOC patients.

While there are increasing reports about the role of bothEPCAM and VCAM-1 in various cancers, only a few reportsof EPCAM and VCAM-1 expression in EOC have beenpublished. The majority of the studies concerning the influ-ences of these CAMs in EOC have been performed in tissuescale in the literature. Therefore, the objective of this studywas to determine the value of circulating EPCAM andVCAM-1 in EOC patients.

Materials and methods

Patients

A total of 50 EOC patients with histologically proven diag-nosis treated at Istanbul University, Institute of Oncologywereenrolled into the study. The staging was established in accor-dance with the International Federation of Gynecologists and

Obstetricians (FIGO) classification. Patients with stage IIIwith bulky disease or stage IV disease were initially treatedwith neoadjuvant chemotherapy and operated afterwards. Thepatients with operable stage III disease who had undergoneprimary surgery including total abdominal hysterectomy, bi-lateral salpingoopherectomy, appendectomy, omentectomy,and pelvic and/or para-aortic lymphadenectomy were treatedwith adjuvant chemotherapy. All patients received standardpaclitaxel–carboplatin-containing chemotherapy regimen.

For comparison of serum levels of EPCAM and VCAM-1,age- and sex-matched 30 healthy controls were included in theanalysis. Informed consent was obtained from all patients, andthe study was reviewed and approved by a local ethicalcommittee.

Measurement of serum EPCAM and VCAM-1 levels

Serum samples were obtained on first admission before anyadjuvant and metastatic treatment was given or before thepatients were followed up. The blood samples were obtainedfrom patients and healthy controls by venipuncture and clottedat room temperature. The sera were collected following cen-trifugation (10 min, 4,000 rpm, at room temperature) andfrozen immediately at −20 °C until analysis.

Serum EPCAM (Hangzhou Eastbiopharm Co., Ltd, China)and VCAM-1 (Bender MedSystems GmbH eBioscience,Austria) levels were determined by the solid-phase sandwichELISA method. The EPCAM ELISA (Eastbiopharm, China)uses a double-antibody sandwich enzyme-linked immunosor-bent assay to determine the level of human EPCAM in thesamples. The serum samples and standards are added to thewells which are pre-coated with human EPCAM monoclonalantibody. Following incubation, the EPCAM antibodies la-beled with biotin and combined with streptavidin–HRP areadded to form immune complex and were allowed to incubatefor 1 h. Unbound material is washed away, and then chromo-gen solution is added for the conversion of the colorlesssolution to a blue solution, the intensity of which is propor-tional to the amount of EPCAM in the sample. As an effect ofthe acidic stop solution, the color has become yellow. Thecolored reaction product is measured using an automatedELISA reader (Rayto, RT-1904C Chemistry Analyzer,Atlanta, GA, USA). The results were expressed in ng/mL.

The VCAM-1 ELISA (eBioscience) uses a double-antibody sandwich enzyme-linked immunosorbent assay todetermine the level of human VCAM-1 in the samples. Theserum samples and standards are added to the wells which arepre-coated with human VCAM monoclonal antibody. Biotin-conjugated anti-human VCAM-1 antibody and streptavidin–HRP are added. Biotin-conjugated anti-human VCAM-1 an-tibody binds to human VCAM-1 captured by the first anti-body. Streptavidin–HRP binds to the biotin-conjugated anti-human VCAM-1 antibody. Following incubation, unbound

Tumor Biol.

streptavidin–HRP is removed during a wash step, and thesubstrate solution reactive with HRP is added to the wells. Acolored product is formed in proportion to the amount ofhuman VCAM-1 present in the sample or standard. Thereaction is terminated by an addition of acid (stop solution),and absorbance is measured using an automated ELISA reader(Rayto, RT-1904C Chemistry Analyzer). The results wereexpressed in ng/mL.

Statistical analysis

Continuous variables were categorized using median valuesas cutoff point. In addition to comparisons of the serumVCAM-1 and EPCAM levels in EOC patients and healthycontrols, the comparison of the patient subgroups according tovarious clinical/laboratory parameters were carried out usingMann–Whitney U test. Overall survival (OS) was calculatedfrom the date of first admission to the clinics to disease-relateddeath or date of last contact with the patient or any familymember. Progression-free survival (PFS) was calculated fromthe date of admission to the date of first radiologic progressionwith/without elevated serum tumor marker. Kaplan–Meiermethod was used for estimation of survival distribution, anddifferences in survival were assessed using log-rank statistics.A p value<0.05 was considered significant. Statistical analy-sis was carried out using SPSS 16.0 software.

Results

A total of 50 patients with a pathologically confirmed diag-nosis of EOC were enrolled into this study. Baseline histo-pathological characteristics and demographic features of pa-tients are listed in Table 1. The median age of the patients was56.5 years, range 22 to 83 years. Majority of the patients hadadvanced disease (FIGO stages III–IV) (90 %).

The levels of serum EPCAM and VCAM-1 in the EOCpatients and healthy controls are shown in Table 2. Thebaseline serum EPCAM levels of the EOC patients weresignificantly higher than those of the control group(3.7 ng/mL vs. 3.0 ng/mL, p =0.03) (Fig. 1a). However, therewas no significant difference in the serum VCAM-1 levelbetween the EOC patients and the controls (948.8 ng/mL vs.769.3 ng/mL, p =0.24) (Fig. 1b).

Table 3 shows the correlation between the serum assaylevels and clinicopathological factors. Metastatic patientshad higher serum VCAM-1 levels compared with the non-metastatic patients (1,443.7 ng/mL vs. 898.1 ng/mL, p =0.03).However, no other clinical variables including response tochemotherapy were found to be correlated with both serumassays (p >0.05). No correlation was found between theserum EPCAM and VCAM-1 levels in EOC patients(r s=0.105, n =50, p =0.362) (Fig. 2).

The median follow-up time was 12 months (range 1–78 months). The median survival for all patients was 52±7.5 months (95 % CI=37–67 months). Overall survival rates(1, 2, and 3 years) were 84.9 % (95 % CI=73.5–96.2), 77 %(95 % CI=62.3–91.7), and 55 % (95 % CI=27–83), respec-tively. Patients with no chemotherapy response and with plat-inum resistance in relapsed disease have significant adverseimpact on both PFS and OS analyses (Table 4). However, nocorrelation was found between serum adhesion molecule as-says and outcome (Table 4, Figs. 3 and 4).

Table 1 Characteristics of the patients and disease

Variables Number

Patients 50

Age of patients (year)

<55 24

≥55 26

Histology

Serous papillary 21

Endometrioid 5

Mixed 4

Clear cell 2

Mucinous 1

Undifferentiated 17

Histological grade

I 2

II 6

III 13

Stage of disease

I–II 5

III 33

IV 12

Debulking surgery (optimal)

Yes 22

No 28

Serum hemoglobin level (12 g/dL)

Low 23

Normal 27

Serum LDH level (480 U/L)

Normal 31

High 13

CA 125 level (35 IU/ml)

Normal 10

High 40

Response to chemotherapy

Yes 25

No 12

Platinum sensitivity in relapse

Sensitive 8

Resistant 15

Tumor Biol.

Discussion

The results of the present study suggest that serum EPCAMlevel was only found to be of diagnostic value. The baselineserum EPCAM levels were significantly higher in EOC pa-tients than those in the control group (p =0.03). However, theserum EPCAM concentration did not correlate with responseto chemotherapy and had no prognostic value on survival.Additionally, no significant associations were found betweenresults of serum EPCAM level and clinicopathological param-eters such as histology, grade, and stage of disease.

Available studies of EPCAM have been limited to paraffin-embedded materials and cell line trials, so all data aboutEPCAM in EOC had been provided by tissue cultures[2–11]. However, our study was performed in serum, for thefirst time in the literature, instead of tissue. So, herein, wediscuss our results with findings that have been provided intissue cultures.

The most striking feature of this study was the demonstrationthat serum EPCAM levels were significantly higher in patientswith EOC than those in healthy controls. Therefore, it shows thatserum EPCAM level has diagnostic value in these patients.Similarly, previous studies have demonstrated that EPCAM isfrequently overexpressed in EOC [2–8]. High-level EPCAMexpression is well documented in the main histotypes of ovariancancer, ranging from 88 to 100 % [2, 3]. Only Brenner tumorshowed lower EPCAM expression (67 %) compared with otherusual histologies such as, endometrioid (88 %), mucinous(100 %), and serous adenocarcinoma (100 %). Spizzo et al.assessed by immunohistochemistry the expression of EPCAMantigen on tissue microarrays containing paraffin-embeddedtissue samples of 199 patients with EOC [6]. They observed arate of EPCAMoverexpression of 68.8%.Histological subtypesassociated with higher EPCAM overexpression rate were undif-ferentiated cancer (100 %), clear cell cancer (92.3 %),endometrioid cancer (82.2 %), serous cancer (67.5 %), andsquamous cell cancer (66.7 %), mixed type (42.9 %), andmucinous cancer (48.8 %). EPCAM overexpression rate wassignificantly higher in grade III tumors (84 %) as compared tothose in grades I and II (63 and 56 %, respectively). No signif-icant association was found between EPCAM expression andtumor stage and/or age at diagnosis. A retrospective study in apopulation-based cohort of 500 ovarian cancer patients exam-ined the variation in expression of 21 biomarkers includingEPCAM in four histological subtypes and at different stages ofdisease (FIGO stages I–III) [2, 4]. Expression levels were deter-mined by tissue microarray (TMA). Only EPCAM showedconsistent high expression across all four ovarian cancer sub-types. Additionally, EPCAM was the most frequent expressedbiomarker across all FIGO stages, slightly above CA 125. Inhigh-grade serous ovarian cancer, a range of EPCAM was verystably expressed at a high rate, across all stages.

The role of EPCAM in ovarian tumor progression is un-clear. One study reported that stage III/IV showed lowerEPCAM expression than stage I [8], while in another study,stage III/IV showed higher EPCAM expression than stage I/II

Table 2 The values of serumEPCAM and VCAM-1 levels inEOC patients and healthy controls

Italicized value is statisticallysignificant (p<0.05)

Assay Patients (n =50) Controls (n =30) p value

Median Range Median Range

Serum EPCAM level (ng/mL) 3.7 0.1–16.4 3.0 0.07–16.8 0.03

Serum VCAM-1 level (ng/mL) 948.8 302.2–42,722.5 769.3 455.8–11,000 0.24

Fig. 1 a The values of serum EPCAM assays in EOC patients andcontrols (p =0.03). b The values of serum VCAM-1 assays in EOCpatients and controls (p =0.24)

Tumor Biol.

disease [7]. Interestingly, EPCAM is also highly expressedin both metastatic and relapsed EOC [2, 5]. In metastaticEOC, staining intensity and the percentage of EPCAMpositive tumor cells were at least as high as those in theprimary EOC, mostly even higher. The high staining in-tensity and high percentage of EPCAM expressing tumorcells are sustained in the patients with relapsed EOC.Moreover, in another study with 19 serous ovarian cancer,EPCAM overexpression was detected in 89 % of theprimary and 84 % of the recurrent tumors (p =1.0); hence,no significant change of the EPCAM expression profilecould be identified overtime [11]. In other words, theEPCAM expression profile appears to remain stable duringthe course from the primary throughout the relapse ofserous ovarian cancer.

Although data are limited to single studies with rather smallnumbers of patients, the accumulated evidence suggests that

Table 3 Results of comparisonsbetween the serum assays andvarious clinical/laboratoryvariables

Variables Assays, median (range)

EPCAM (ng/mL) VCAM-1 (ng/mL)

Age, years (p) 0.56 0.18

Young (<55) 3.7 (0.7–16.4) 1210.1 (371.4–2,235.6)

Older (≥55) 3.8 (0.1–13.7) 881.5 (302.2–42,722.5)

Histology (p) 0.43 0.28

Serous papillary 4.2 (2.2–13.7) 973.2 (461.3–1,861)

Others 3.5 (0.1–16.4) 924.4 (30.2–42,722.5)

Grade (p) 0.41 0.60

I+II 3.6 (0.7–7.3) 899.5 (371.4–1,456.7)

III 4.2 (0.1–13.7) 793.1 (505–1,861)

Stage (p) 0.94 0.03

Non-metastatic 3.7 (0.7–16.4) 898.1 (371.4–42 ,722.5)

Metastatic 3.7 (0.1–12.2) 1 ,443.7 (302.2–23,589.5)

Debulking surgery (p) 0.95 0.53

Yes 3.9 (0.7–13.7) 917.8 (371.4–2,051.5)

No 3.6 (0.1–16.4) 1,021.1 (302.2–42,722.5)

Hemoglobin level (p) 0.56 0.99

Low 3.6 (2.2–16.4) 924.4 (371.4–42,722.5)

Normal 3.8 (0.1–12.2) 1,062.4 (302.2–5,171.4)

Serum LDH level (p) 0.26 0.55

Normal 3.6 (0.1–16.4) 913 (371.4–23,589.5)

High 3.8 (2.9–12.2) 1,062.4 (302.2–42,722.5)

Serum CA 125 level (p) 0.59 0.71

Normal 3.5 (2.8–7.3) 909.4 (461.3–2,000.9)

High 3.8 (2.2–13.7) 1,065.8 (302.2–42,722.5)

Response to chemotherapy (p) 0.34 0.58

Yes 3.8 (0.1–16.4) 973.2 (371.4–5,171.4)

No 3.7 (3.2–12.2) 1,157.1 (302.2–42,722.5)

Platinum sensitivity in relapse (p) 0.83 0.77

Sensitive 4.2 (0.1–16.4) 1,085.4 (463.9–2,356)

Resistant 3.6 (2.2–12.2) 1,069.1 (302.2–42,722.5)

Fig. 2 Correlation between serum EPCAM and VCAM-1 levels in EOCpatients (rs=0.105, n =50, p =0.362)

Tumor Biol.

EPCAM is a stable antigen, and its expression in metastasesand relapsed disease correlates with the primary ovarian cancertumor cells. This would be an important prerequisite for ther-apeutic strategies targeting EPCAM [2]. These results indicate

that the majority of EOC cell lines overexpress EPCAM.These data showed that the serum soluble EPCAM was posi-tively correlated with the tissue concentration in EOC patients.

The overexpression of EPCAM was associated with anadvanced stage of disease and linked to worse outcome incertain tumor types [2]. For example, EPCAM expression wasfound to be associated with poor prognosis in patients withbreast cancer, triple-negative breast cancer, prostate cancer,and gallbladder cancer [2]. In contrast, EPCAM expressionwas associated with improved survival in patients with renalcell carcinoma, moderately differentiated stage II colon can-cer, and NSCLC at stage pT2. No significant difference wasfound in disease-free or overall survival between EPCAM-

Table 4 Univariate analyses of progression-free and overall survivals

Variables Survival (month)

Progression-freesurvival, median(±SE)

Overallsurvival,median (±SE)

Age, years (p) 0.60 0.12

Young (<55) 29.3 (9.9) 61.2 (8.9)

Older (≥55) 12.6 (2.6) 24.1 (2.3)

Histology (p) 0.85 0.74

Serous papillary 10.4 (2.2) 26.8 (2.6)

Others 27.3 (7.1) 52.6 (9.5)

Grade (p) 0.36 0.16

I+II NR NR

III NR NR

Stage (p) 0.76 0.34

Non-metastatic 25.0 (7.0) 55.4 (8.3)

Metastatic 14.5 (3.6) 24.2 (4.3)

Debulking surgery (p) 0.47 0.09

Yes 26.1 (6.3) 45.4 (9.0)

No 9.8 (3.7) 30.3 (1.7)

Hemoglobin level (p) 0.80 0.21

Low 14.1 (3.2) 28.2 (1.8)

Normal 23.8 (7.8) 52.3 (8.6)

Serum LDH level (p) 0.11 0.01

Normal 34.5 (7.0) 62.9 (7.6)

High 9.6 (3.8) 21.7 (4.1)

Serum CA 125 level (p) 0.45 0.18

Normal NR NR

High NR NR

Response to chemotherapy (p) <0.001 <0.001

Yes 34.2 (7.0) 60.2 (7.9)

No 3.3 (0.5) 14.5 (4.5)

Platinum sensitivity in relapse(p)

0.001 0.003

Sensitive 46.2 (5.3) 66.0 (10.4)

Resistant 6.9 (1.5) 15.1 (3.2)

Serum EPCAM level (p) 0.71 0.70

High (>median) 14.3 (3.3) 26.8 (2.3)

Low (<median) 21.7 (7.3) 47.0 (9.5)

Serum VCAM-1 level (p) 0.67 0.62

High (>median) 13.1 (2.8) 26.1 (2.5)

Low (<median) 26.4 (8.9) 61.9 (7.3)

Italicized values are statictically significant (p<0.05)

NR Not reached

Fig. 3 Overall survival curves in EOC patients according to EPCAMlevels (p =0.70)

Fig. 4 Overall survival curves in EOC patients according toVCAM-1levels (p =0.62)

Tumor Biol.

overexpressing and EPCAM-negative groups. EPCAM expres-sion has no prognostic significance in those cancers. EPCAMoverexpression in ovarian cancer was found to be particularlystrong in advanced stages of disease [2, 6]. The overexpressionof EPCAM was an independent prognostic marker for reducedsurvival of patients with EOC [6, 9]. Spizzo et al. found thatEPCAM overexpression was significantly related to a decreasedoverall survival (p=0.036) [6]. Its prognostic power was partic-ularly strong in patients with stage III and IV disease. The overallsurvival was twofold higher in patients where EPCAM was notoverexpressed compared to that of patients with EPCAM over-expression (p<0.01). Cox regression analysis showed thatEPCAM overexpression to be an independent prognosticmarker (p =0.037). This retrospective analysis resulted, inthe first time, an independent prognostic value of EPCAMoverexpression in patients with EOC. Later, Shim et al. alsoshowed similar result [9]. They found that the overexpressionof EPCAMcorrelatedwith poor survival in 72EOCpatients (p=0.036). Likewise, in another study, the downregulation ofEPCAM expression resulted to an effect on cell growth inovarian cancer, and the induction or upregulation of EPCAMexpression decreased ovarian cancer survival in ovarian cell lines[10]. Despite these promising results, this could not be confirmedby other independent studies [7, 8]. EPCAM expression was notassociated with survival in these trials. Thus, EPCAMmay serveas an additional prognostic marker in these patients.

Although the high expression of VCAM-1 has been shown tobe associated with several cancers, and increased expression hasbeen found to be related to progression of several cancers, its rolein ovarian cancer is largely undefined [12–14]. Similar to othertumors, the expression of VCAM-1 has been also demonstratedhistologically in tumor cell lines of EOC patients [15, 16]. In arecent study, VCAM-1 expression was examined via IHC stain-ing of 251 high-grade serous carcinoma samples using tissuemicroarray [15]. VCAM-1 positive staining was predominantlylocalized not only in tumor cells but also in stromal cells. Thehigh expression of VCAM-1 was found in the ovarian tumor(40 %) compared with serous cystadenoma (10 %) and normalovarian tissue (0 %) (p=0.02). The overexpression of VCAM-1in EOC cells was associated with advanced age at diag-nosis (p =0.008) and poor response to chemotherapy (p =0.025). Conversely, no statistically significant correlationswere found between VCAM-1 expression and the other clin-icopathologic variables including FIGO stage and serum CA125 level. Clinical data also demonstrated that VCAM-1overexpression was associated with poor prognosis. Thisstudy provides strong evidence that VCAM-1 plays an impor-tant role in ovarian tumor growth, and it may be used as aprognostic factor.

Soluble variants of VCAM-1 molecule in the circulationhave been reported in 1992 [22]. The next year, Banks et al.examined for the first time the concentration of VCAM-1 in aconsecutive series of 110 cancer patients [23]. The soluble

VCAM-1 levels were found to be significantly higher in allthe patient groups compared with those in the controls. So far,elevated serum levels of VCAM-1 have been identified invarious malignant diseases, including breast cancer [17, 18]and colorectal cancer [19–21]. In these studies, solubleVCAM-1 levels in patients were significantly higher thanthose in the control subjects. The increased soluble levels ofVCAM-1were associated with advanced stage and prognosticsignificance. For ovarian cancer, as the first study, the solubleVCAM-1 levels were found to be significantly higher in 15ovarian cancer patients compared with those in the controls,similar to other tumors [23]. In another study, the concentra-tion of serum VCAM-1 was assayed in 67 EOC cases usingELISA [24]. The serum VCAM-1 level in EOC was higherthan those in benign ovarian tumor and normal cases(p <0.01). The serum VCAM-1 levels in stages II–IV werehigher than that in stage I (p <0.05). Likewise, the patientswith grade III EOC had significantly elevated levels thanpatients with grades I and II (p <0.01). Moreover, no correla-tion was found between serum VCAM-1 levels and histolog-ical subtypes (p >0.05). They demonstrated that serumVCAM-1 level was not correlated with overall survival. Onthe other side, we found that baseline serum VCAM-1 levelswere higher than those in the control group, but the p valuewas insignificant. Although tumor histology and grade werenot correlated with serum VCAM-1 assay, elevated serumVCAM-1 levels was associated with advanced stage. Ourfindings are consistent with the data regarding no prognosticmarker of serum VCAM-1 level. These data showed thatserum soluble VCAM-1 level was positively correlated withthe tissue concentration in EOC patients.

The differences in the findings with regard to EPCAM andVCAM-1 probably reflect the differences in source, kineticsof expression or destruction, and possibly signals includingtheir expression and release [23]. Additionally, these contra-dicting results of these several studies might be attributable toother several factors. So far, no consensus exists on whichtumors and methods should be used for testing EPCAM andVCAM-1 expressions. During the recent decades, IHC hasbecome a useful adjunctive method in diagnostic histopathol-ogy. There are a number of drawbacks with IHC, the mostimportant of which are lack of assay standardization andvariance in the interpretation of the IHC staining. In manycases, the studies were performed on a relatively small samplesize, which may have been insufficient to show significantdifferences. A standardized method remains to be establishedand validated in larger series of patients in prospective studies.

In conclusion, EPCAM level in serum was found to be adiagnostic marker similar to tissue investigation in EOC pa-tients. In addition, serum VCAM-1 level has a diagnostic rolefor differentiating metastatic disease from early disease.Neither serum EPCAM level nor VCAM-1 level has predic-tive and prognostic value in EOC patients. Although this

Tumor Biol.

study has its limitations due to limited patient numbers, thepresent study also contributes to the literature. Further, morelarge-scale studies are needed to determine the exact role ofserum EPCAM and VCAM-1 levels in terms of clinicalsignificance in EOC patients.

Conflicts of interest None

References

1. Charalabopoulos K, Papalimneou V, Charalabopoulos A. Adhesionmolecules in lung cancer. Exp Oncol. 2003;25:16–21.

2. Patriarca C, Macchi RM, Marschner AK, Mellstedt H. Epithelial celladhesion molecule expression (CD326) in cancer: a short review.Cancer Treat Rev. 2012;38:68–75.

3. Went PT, Lugli A, Meier S, Bundi M, Mirlacher M, Sauter G, et al.Frequent EpCAM protein expression in human carcinomas. HumPathol. 2004;35:122–8.

4. Kobel M, Kalloger SE, Boyd N, McKinney S, Mehl E, Palmer C,et al. Ovarian carcinoma subtypes are different diseases: implicationsfor biomarker studies. PLoS Med. 2008;5:e232.

5. Bellone S, Siegel ER, Cocco E, Cargnelutti M, Silasi DA, Azodi M,et al. Overexpression of epithelial cell adhesion molecule in primary,metastatic, and recurrent/chemotherapy-resistant epithelial ovariancancer: implications for epithelial cell adhesion molecule-specificimmunotherapy. Int J Gynecol Cancer. 2009;19:860–6.

6. Spizzo G, Went P, Dirnhofer S, Obrist P, Moch H, Baeuerle PA, et al.Overexpression of epithelial cell adhesion molecule (Ep-CAM) is anindependent prognostic marker for reduced survival of patients withepithelial ovarian cancer. Gynecol Oncol. 2006;103:483–8.

7. Heinzelmann-Schwarz VA, Gardiner-GardenM, Henshall SM, ScurryJ, Scolyer RA, Davies MJ, et al. Overexpression of the cell adhesionmolecules DDR1, Claudin 3, and Ep-CAM in metastatic ovarianepithelium and ovarian cancer. Clin Cancer Res. 2004;10:4427–36.

8. Kim JH, Herlyn D, Wong KK, Park DC, Schorge JO, Lu KH, et al.Identification of epithelial cell adhesion molecule autoantibody inpatients with ovarian cancer. Clin Cancer Res. 2003;9:4782–91.

9. Shim HS, Yoon BS, Cho NH. Prognostic significance of pairedepithelial cell adhesion molecule and E-cadherin in ovarian serouscarcinoma. Human Pathol. 2009;40:693–8.

10. Van der Gun BT, Huisman C, Stolzenburg S, Kazemier HG, RuitersMHJ, Blancafort P, et al. Bidirectional modulation of endogenousEpCAM expression to unravel its function in ovarian cancer. Br JCancer. 2013;108:881–6.

11. Pietzner K, Woopen H, Richter R, Joens T, Braicu EI, Dimitrova D,

et al. Expression of epithelial cell adhesion molecule in paired tumorsamples of patients with primary and recurrent serous ovarian cancer.Int J Gynecol Cancer. 2013;23:797–802.

12. Heidemann J,Maaser C, Lugering A, Spahn TW, Zimmer KP, HerbstH, et al. Expression of vascular cell adhesion molecule-1 (CD 106) innormal and neoplastic human esophageal squamous epithelium. Int JOncol. 2006;28:77–85.

13. Hemmerlein B, Scherbening J, Kugler A, Radzun HJ. Expression ofVCAM-1, ICAM-11E- and P-selectin and tumour-associated macro-phages in renal cell carcinoma. Histopathology. 2000;37:78–83.

14. den Brekel AJ S-v, Thunnissen FB, Buurman WA, Wouters EF.Expression of E-selectin, intercellular adhesion molecule (ICAM)-1and vascular cell adhesion molecule (VCAM)-1 in non-small-celllung carcinoma. Virchows Arch. 1996;428:21–7.

15. Huang J, Zhang J, Li H, Lu Z, Shan W, Mercado-Uribe I, et al.VCAM1 expression correlated with tumorigenesis and poor progno-sis in high grade serous ovarian cancer. Am J Transl Res. 2013;5:336–46.

16. Slack-Davis JK, Atkins KA, Harrer C, Hershey ED, Conaway M.Vascular cell adhesion molecule-1 is a regulator of ovarian cancerperitoneal metastasis. Cancer Res. 2009;69:1469–76.

17. O'Hanlon DM, Fitzsimons H, Lynch J, Tormey S, Malone C, GivenHF. Soluble adhesion molecules (E-selectin, ICAM-1 and VCAM-1)in breast carcinoma. Eur J Cancer. 2002;38:2252–7.

18. Silva HC, Garcao F, Coutinho EC, De Oliveira CF, Regateiro FJ.Soluble VCAM-1 and E-selectin in breast cancer: relationship withstaging and with the detection of circulating cancer cells. Neoplasma.2006;53:538–43.

19. Karayiannakis AJ, Syrigos KN, Zbar A, Baibas N, Polychronidis A,Simopoulos C, et al. Clinical significance of preoperative serumvascular endothelial growth factor levels in patients with colorectalcancer and the effect of tumor surgery. Surgery. 2002;131:548–55.

20. Chin KF, Greenman J, Gardiner F, Kumar H, Topping K, Monson J.Pre-operative serum vascular endothelial growth factor can selectpatients for adjuvant treatment after curative resection in colorectalcancer. Br J Cancer. 2000;83:1425–31.

21. Kemik O, Kemik AS, Hasirci I, Adas M, Purisa S, Dulger AC, et al.Serum level of soluble vascular adhesion molecule 1 in patients withrectal cancer. Eur J Gen Med. 2011;8:105–9.

22. Gearing AJH, Hemmingway IK, Pigott R, Hughes J, Rees AJ,Cashman SJ. Soluble forms of vascular adhesion molecules, E-selectin, ICAM-1 and VCAM-1: pathological significance. AnnNYAcad Sci. 1992;667:324–31.

23. Banks RE, Gearing AJH, Hemingway IK, Norfolk DR, Perren TJ,Selby PJ. Circulating intercellular adhesionmolecule-1 (ICAM-1), E-selectin and vascular cell adhesion molecule-1 in human malignan-cies. Br J Cancer. 1993;68:122–4.

24. Qiao XM,Wang ZM. Significance of concentration of serum solublevascular cell adhesion molecule-1 in epithelial ovarian carcinoma. AiZheng. 2004;23:81–4.

Tumor Biol.