POLSKI PRZEGLĄD CHIRURGICZNY 10.2478/v10035-012-0057-52012, 84, 7, 341–347

vAScuLAR ENDOthELIAL GROwth fActOR AND SuRvIvIN ImmuNOStAINING IN GAStRIc ADENOcARcINOmA

jARoSłAw BuRy1, juSTynA SzuMiło2, AndRzej dąBRowSKi3, AleKSAndeR ciechAńSKi3, juSTynA ŚliwińSKA2, GrzEGorz WAllnEr3

Department of General Surgery, Hospital of Ministry of Interior in Lublin1 Kierownik: lek. E. bury

Department of Clinical Pathomorphology, Medical University in Lublin2 Kierownik: dr hab. j. Szumiło

2nd Department of General, Gastrointestinal and Oncological Surgery of the Alimentary Tract, Medical University in Lublin3

Kierownik: prof. dr hab. G. Wallner

Two molecules – vascular endothelial growth factor involved in new vessels formation and survivin – antiapoptotic protein, reported to be associated with worse prognosis in various malignancies have been chosen for the study. Both are potential target for novel therapiesthe aim of the study was to determine the immunostaining of VEGF and survivin in gastric carci-noma and to analyse their relationship to the selected clinicopathological features and survival.material and methods. Formalin-fixed, paraffin-embedded sections from 41 gastric adenocarcinomas were used for immunohistochemical reaction with monoclonal antibodies against vascular endothelial growth factor and survivin. The results were compared with selected clinicopathological features and survival. Results. Positive immunohistochemical reaction for vascular endothelial growth factor and survivin was revealed in 24 (58,53%) and 30 (73,17%), gastric carcinomas respectively. Vascular endothelial growth factor-negative gastric carcinomas were significantly more common in cases without metasta-ses to regional lymph nodes and distant organs and in less advanced cases. Similar, distant metasta-ses were also statistically less common in survivin-negative carcinomas. The differences in immuno-histochemical reactions for survivin between less and more advanced cases almost reach statistical significance. The only factors significantly influenced 1, 2 and 3-year survival were vascular endothe-lial growth factor and survivin status. Statistically significant higher percentage of survival was noted in patients with vascular endothelial growth factor- and survivin-negative tumors.conclusions. It seems that vascular endothelial growth factor and survivin play role in local invasion and spread of gastric adenocarcinoma and negatively influences survival. However, further studies are required to assess their true usefulness in the clinical practice. Key words: vascular endothelial growth factor, VEGF, survivin, adenocarcinoma, stomach, survival

Gastric carcinoma is one of the most com-mon malignancies in the world. Despite the great progress in its early detection and ther-apy the prognosis is still poor with a 5-year survival rate of 20-25% (1). The only poten-tially curative treatment for locally advanced gastric carcinoma is complete surgical resec-tion, now supplied by modern minimally inva-sive techniques. Since lymph nodes metastases

are common event that negatively influence prognosis, adjuvant and neoadjuvant therapies should be considered as a therapeutic option (1). Most recently many novel targeted thera-pies are introduced like cell cycle inhibitors, apoptosis promoters, antiangiogenic agents, matrix metalloproteinases inhibitors etc. They require detailed understanding of biol-ogy of the tumor, especially mechanisms in-

Brought to you by | Ryerson UniversityAuthenticated | 141.117.79.62

Download Date | 4/28/13 11:55 AM

342 J. Bury et al.

volved in invasiveness and metastases forma-tion (2, 3).

Vascular endothelial growth factor (VEGF) is a protein secreted by almost all tumors that influences endothelium and promotes prolif-eration of new vessels by binding specific recep-tors (Flt-1 and KDR). Current reports proved that over-expression of VEGF in malignant tumors are associated with tumor growth, metastasis and worse prognosis (3, 4). Anti-angiogenic therapy has been introduced in some digestive tract malignancies, mostly in colorectal carcinomas, but in the stomach it is still experimental (5).

Among intracellular apoptosis inhibitors, survivin is the most important protein in the light of diagnostics and therapy. Apart from hampering apoptosis, survivin plays role in proliferation. Over-production of survivin and its mRNA has been observed in many malig-nant tumors including carcinomas of the lung, breast, stomach, colon and ovary as well as melanoma and lymphomas. It is associated with worse therapy results, higher risk of re-currences and resistance to chemotherapy (2, 6). Since apoptosis inhibitors are present mainly in cancerous cells they are excellent goal for anti-neoplastic therapy.

The aim of the present study was to deter-mine the immunostaining of VEGF and sur-vivin in gastric carcinoma and to analyse their relationship to the selected clinicopathological features and survival.

MATERIAL AND METHODS

The study was conducted on 41 patients with gastric carcinoma admitted to the 2nd Department of General Surgery, Medical Uni-versity in Lublin. They were treated with total or sub-total gastrectomy with at least D2 lymphadenectomy. The study group consisted of 11 females (26.83%) and 30 males (73.17%). The age of the patients ranged from 23 to 73 years (mean 61,12; median 64 years). The preoperative staging of the tumor was as-sessed based on: physical examination, gas-troscopy, ultrasound examination, chest ra-diogram and in some cases computed tomog-raphy or endoscopic ultrasonography. Local invasion was evaluated during intraoperative assessment and subsequently supported by pathological examination. The staging of the

tumors was as follow: 5 patients (12.3%) in stage IA, 8 patients (19.6%) in IA, 7 patients (17.09%) – stage II; 3 patients – stage IIIA, 1 patient – stage IIIB and 17 patients (41.5%) – stage IV. Lymph node metastases were detected in 26 patients while distant metas-tases were seen in 16 patients. The observa-tion periods were dived into three models: 12, 24 and 36 months. Fifteen patients (36.59%) died before 12 months with one patient dying in the early postoperative period. After 24 months the percentage of deaths was 58.54% (24 patients) and after 36 months – 78.05% (32 patients).

The surgical specimens were fixed in 10% buffered formalin. The samples were taken from the tumor and from normal-appearing wall of the stomach and then divided into 3-7 slides. In every case the proximal and distal resection margins and all lymph nodes were examined. The sections were routinely pro-cessed, embedded in paraffin blocks and stained with hematoxylin and eosin, mucicar-mine, PAS and alcian blue, high iron diamine and Warthin-Starry methods. Evaluation of pathologic stage of the disease was based on the TNM system and the histological subtype of the carcinoma on the World Health Organi-zation’s criteria and Lauren classification (7).

Immunohistochemical study was performed on 4-µm sections from one, representative paraffin block from each primary gastric tu-mor. Before the main reaction with survivin antibody was performed, antigen retrieval technique was applied and slices were pre-treated in citrate buffer (0.01 M, pH 6.0) in a microwave oven (3x for 5 min). The slides for VEGF were pretreated by submerging in Tar-get Retrieval Solution (pH 9.0; Dako, Den-mark) and then placing in water at 95oC for 20 minutes. For all antigens the endogenous peroxidase activity was blocked by 3% hydro-gen peroxide solution for 5 min at the room temperature. The samples for survivin detec-tion were then washed with TBS and 0.1% Tween20 solution three times for 5 minutes and afterwards washed with horse serum for 20 minutes. Next step was to use the primary antibody – monoclonal mouse antibody against human survivin (clone 12C4, Dako) in dilution 1:25 for 60 minutes. The slides were then washed 3 times for 5 minutes with TBS and incubated with ImmPressTM Universal Re-

Brought to you by | Ryerson UniversityAuthenticated | 141.117.79.62

Download Date | 4/28/13 11:55 AM

343Vascular endothelial growth factor and survivin immunostaining in gastric adenocarcinoma

agent/Peroxidase (Vector Laboratories, USA) for 30 minutes. On the other hand, VEGF samples, after washing, were incubated with monoclonal mouse antibody against human VEGF (clone VG1, Dako) in dilution 1:20 for 60 minutes and then with complex DakoEnvi-sion+/HRP, Mouse (Dako). All incubations were performed at room temperature and TBS was used every time the slides were washed. The specific immune reactions were visualized by 3’,3-diaminobenzidine tetrahydrochloride (DAB) (Dako) for about 2 minutes in room temperature. Sections were than counter-stained with Mayer’s haematoxylin. Normal colonic mucosa was applied as a positive con-trol. Slides treated in the same way, but with replacement of primary antibodies, were used as negative controls. Immunohistochemical reaction was evaluated by light microscope (Olympus BX45; Japan) equipped with eye grid. Percentage of positively-stained cells and the type of reaction (nuclear or cytoplasmic reaction) of each category were calculated in approximately 1000 cells of the some type in randomly selected areas using magnification 200x (4-7 fields).

Statistical differences between variables were evaluated using t-Student test or U Mann-Whitney and one-way analysis of vari-

ance was done using Kruskal-Wallis ANOVA. The quality parameters were analyzed by χ2 and Fisher χ2 test. Survival rates were calcu-lated using Cox, χ2 and log-rank tests. P values less than 0.05 were considered significant. Data were analysed by STATISTICA 5.0 on IBM compatible computer.

RESULTS

Positive immunohistochemical reaction for VEGF and survivin was revealed in 24 (58.53%) and 30 gastric carcinomas (73.17%), respectively. For VEGF the reaction was ex-clusively cytoplasic (fig. 1A), whereas for sur-viving both cytoplasmic and nuclear (fig. 1B).

VEGF-negative gastric carcinomas were significantly more common in cases without metastases to regional lymph nodes and dis-tant organs and in less advanced cases (Table 1). The differences between other evaluated features, i.e., depth of infiltration and histo-logical subtypes according to WHO and Lauren classifications were not significant.

No statistically significant differences were noted between survivin-positive and -negative cases and depth of stomach wall infiltration, regional lymph nodes status as well as histo-

Table 1. Results of immunohistochemical reaction for VEGF and selected clinicopathological features in patients with gastric adenocarcionoma

(-) (+) pn % n %pT1 4 23,53 1 4,17 0120pT2 10 58,82 6 25pT3 3 17,65 12 50pT4 0 0 6 20,83pN0 10 58,82 5 20,83 < 0,01pN1 5 29,41 4 16,67pN2 1 5,88 10 41,67pN3 1 5,88 5 20,83M0 13 76,47 12 50 < 0,05M1 3 23,53 13 50 I, II 12 70,59 8 33,33 < 0,05 III, IV 5 29,41 16 66,67Tubular adenocarcinoma 11 100 21 70

0,120Signet ring cell carcinoma 0 0 7 23,33Mucinous adenocarcinoma 0 0 2 6,67Intestinal 11 64,71 9 37,50

0,157Diffuse 0 0 2 8,33Mixed 6 35,29 13 54,17(-), (+) – negative and positive immunochistochemical reaction in gastric adenocarcinoma; staging according to pTNM (7); p – poziom istotności statystycznej / level of significance

Brought to you by | Ryerson UniversityAuthenticated | 141.117.79.62

Download Date | 4/28/13 11:55 AM

344 J. Bury et al.

logical subtypes of carcinoma (tab. 2). How-ever, distant metastases were statistically less common in survivin-negative carcinomas. The differences in immunohistochemical reaction between less and more advanced cases almost reach statistical significance.

The only factors significantly influenced 1, 2 and 3-year survival in patients with gastric carcinoma were VEGF and surviving status.

Statistically significant higher percentage of survival was noted in patients with VEGF- (fig. 1A) and survivin-negative tumors (fig. 2B).

DISCUSSION

Many studies proved that VEGF and sur-vivin are proteins produced by variety of ma-

Fig. 1. Positive immunohistochemical reaction for VEGF (A) and survivin (B) in gastric tubular adenocarcinoma (objective magn. x10)

A B

Table 2. Results of immunohistochemical reaction for survivin and selected clinicopathological features in patients with gastric adenocarcionoma

(-) (+) pn % n %pT1 2 18,18 3 10 0,178pT2 6 54,55 10 33,33pT3 1 9,09 14 46,67pT4 2 18,18 3 10pN0 6 54,55 9 30 0,120pN1 3 27,27 6 20pN2 0 0 11 36,33pN3 2 18,18 4 13,33M0 10 90,91 15 50 < 0,05M1 1 9,09 15 50I, II 8 72,73 12 40 0,065III, IV 3 27,27 18 60Tubular adenocarcinoma 14 82,35 18 75,00 0,474Signet ring cell carcinoma 3 17,65 4 16,67Mucinous adenocarcinoma 0 0 2 8,33Intestinal / 7 63,64 13 43,33 0,419Diffuse 0 0 2 6,67Mixed 4 36,36 15 50(-), (+) – ujemna i dodatnia reakcja immunohistochemiczna u pacjentów z gruczolakorakiem żołądka; ocena stopnia zaawansowania zgodnie z klasyfikacją pTNM (7) / negative and positive immunochistochemical reaction in gastric adenocarcinoma; staging according to pTNM (7); p – poziom istotności statystycznej / level of significance

Brought to you by | Ryerson UniversityAuthenticated | 141.117.79.62

Download Date | 4/28/13 11:55 AM

345Vascular endothelial growth factor and survivin immunostaining in gastric adenocarcinoma

Fig. 2. Three-year survival of patients with gastric adenocarcinoma according to VEGF (A) and survivin (B) status

A B

lignancies including gastric, pulmonary, esophageal, breast, colon, hepatic, pancreatic, urothelial, nasopharyngeal, uterine and ovar-ian carcinomas. The obtained data suggests that expression of both of these factors has a negative influence on prognosis (3, 6, 8).

In the presented study, the role of both molecules was assessed according to their in-fluence on the survival rate of patients with gastric adenocarcinoma. In the study group, VEGF was detected in 24 patients (58.53%) and survivn in 30 out of 41 cases (73.17%). Similar VEGF rate was obtain in other reports – Li et al. (6) detected VEGF in 76% and Choi et al. (9) in 62% of patients. Like in our study lack of statistically significant differences be-tween survivn and VEGF status and age or histological type of the tumor was revealed in esophageal, colonic and nasopharyngeal car-cinomas (10, 11). Statistically significant con-nection was found between survivin expression and distant metastasis and between VEGF and the nodal and distal metastasis. Other studies report resembling data. Lu et al. (12) described significantly lower apoptotic index and co-expression of factors regulating apop-tosis like p53 and bcl-2 in tumors secreting survivin. VEGF and its receptor VEGFR-3 were researched by others and proven to have connection with lymph node and distant me-tastasis in 91 patients (13). In addition, sig-nificantly lower survival rate was observed in patients with higher VEGF expression. Sun et al. (4) found association between expression of survivin and VEGF and lymph node metasta-sis, pT and staging according to TNM UICC.

Connection between survivin expression and distant metastasis was also noted.

In the following part of our research, the results were compared with the staging accord-ing to TNM-UICC classification. There was a strong connection between VEGF immunos-taining and the advanced stages of the tumor (p<0.05) and close to statistical significance in case of survivin immunostaining (p=0.065). Majority of studies also reports connection between the stage and VEGF immunostaining (14). Similar data pertains not only to gastric malignancy (15) but also to esophageal (16) and colonic cancers (17). Song et al. (15) indi-cated that presence of VEGF in tumor cells has a strong connection with metastasis to re-gional lymph nodes and the stage according to TNM UICC classification.

In our study, survivin immunostaining was compared to the mean survival time and it was 17.05 months compared to 32.55 months in patients with negative reaction for surviving (p<0.05). Similar results were obtained com-paring VEGF, where the survival was signifi-cantly shorter in patients with positive reac-tion: subsequently 14.52 and 30.65 months (p<0.05). Patients with co-expression of both proteins had significantly lower survival time (p<0.0005). Similar data has been reported from other clinical centers (18). From 1-, 2- and 3- year observation a strong correlation be-tween survivin and VEGF status and survival time was revealed. Statistically significant higher percentage of survival was noted in patients with VEGF- and survivin-negative tumors.

Brought to you by | Ryerson UniversityAuthenticated | 141.117.79.62

Download Date | 4/28/13 11:55 AM

346 J. Bury et al.

The golden standard in treatment of gastric carcinoma is surgical procedure. Although the procedures are more oncologically radical, the result is still not satisfying and requires ad-ditional treatment. The results of presented research suggest a significant role of survivin and VEGF in biology of gastric carcinoma. Nowadays, studies on new targeted therapies focused on those two proteins are in progress. The study of anti-angiogenic therapy is also advanced and various chemicals were applied, e.g. the anti-VEGF receptor antibodies (19), anti-VEGF antibodies (Bevacizumab) (20), peptides inhibiting the activity of VEGFR-2, VEGFR tyrosine kinase inhibitors and oligo-nucleotides hampering the expression of VEGF gene (21). Research pertaining to pulmonary, breast and colonic cancer as well as melanoma suggests that they significantly reduce the development of the tumor and formation of metastases in lung carcinoma (22).

It seems that the therapy hampering VEGF and survivin expression is not only efficient but also has minimal toxic effect on the healthy tissues. This insignificant toxicity may be con-nected with lower expression of survivin in normal cells vs. neoplastic tissue or the time of observation was not long enough for the negative effect to become evident. Potential side effects of anti-survivin therapy may per-tain to proliferating cells such as hematopoi-etic cell, lymphocytes T, endothelial cells lining the gastro-intestinal tract, hepatocytes and neurons. Particularly the action on lympho-

cytes T seems worrisome as they are the main mean eliminating the neoplastic cells (23). Other cases of proven side effects embrace bone marrow stem cells, however the toxicity was minimal compared to overall inhibition of survivin in tumor cells (24).

Another opportunity which is provided by the fact that the expression of survivin is dif-ferent in normal and tumor cells is using it as early marker of carcinogenesis. This kind of research has been conducted in urinary blad-der and oral malignancies and in the latter elevated level of survivin was detected even in precancerous lesions (25, 26). Similar data indicated increase in survivin expression also in colonic and breast adenomas and Bowen disease which suggests that expression of sur-vivin may be present during malignant trans-formation or be a sign of imbalance in apopto-sis and cellular cycle. Summarizing, survivin may be detected in samples of bodily fluids, easily obtained form the patient and may identify not only tumors but lesions with the increased risk of malignant transformation (26, 27).

Although the results of this and other stud-ies on survivin and VEGF expression/immu-nostaining with their clinical meanings seem to be interesting, the real value at this stage may only be in addition to other methods. Pos-sible prognostic value of both proteins may be promising nevertheless further studies are required to assess the true usefulness in the clinical practice.

REFERENCES

1. meyer hJ, Wilke h: Treatment strategies in gastric cancer. dtsch Arztelb int 2011; 108: 698-706. 2. Wang tt, Qian XP, liu br: Survivin: potential role in diagnosis, prognosis and targeted therapy of gastric cancer. World J Gastroenterol 2007; 13: 2784-90.3. Yasiu W, oue n, Kuniyasu h: Molecular diagno-sis of gastric cancer: present and future. Gastr Cancer 2001; 4: 113-21.4. sun Ys, Ye zY, zhao zs: Expression of vascu-lar endothelial growth factor C and survivin in gastric carcinoma and their clinical implications. zhonghua Wei Chang Wai Ke za zhi 2006, 9: 264-67.5. iordache s, saftoiu A, Georgescu CV et al.: Va-scular endothelial growth factor expression and

microvessel density – two useful tools for the as-sessment of prognosis and survival in gastric can-cer patients. J Gastrointestin liver dis 2010; 19: 135-39.6. li Yh, Wang C, meng K: Influence of survivin and caspase-3 on cell apoptosis and prognosis in gastric carcinoma. World J Gastroenterol 2004; 10: 1984-88.7. lauwers Gy, carneiro F, Graham dy et al.: Curado MP: Gastric carcinoma. W: Bosman FT, Carneiro F, Hruban RH, Theise ND (red.) WHO Classification of tumors of digestive system. Lyon: IARC 2010; s. 45-80.8. Wang zn, Xu hm, Jiang l: Expression of survi-vin in primary and metastatic gastric cancer cells obtained by laser capture microdissection. World J Gastroenterol 2004; 10: 3094-98.

Brought to you by | Ryerson UniversityAuthenticated | 141.117.79.62

Download Date | 4/28/13 11:55 AM

347Vascular endothelial growth factor and survivin immunostaining in gastric adenocarcinoma

9. Choi Jh, oh Yh: Correlation of Vascular Endo-thelial Growth Factor-D expression and VEGFR-3-positive vessel density with lymph node metasta-sis in gastric carcinoma. J Kor med sci 2008; 23: 592-97.10. dąbrowski A, Filip A, zgodziński w et al.: As-sessment of prognostic significance of cytoplasmic survivin expression in advanced oesophageal can-cer. Folia histochem cytobiol 2004; 42: 169-72.11. li yh, hu cF, Shao Q: Elevated expressions of survivin and VEGF protein are strong indepen-dent predictors of survival in advanced nasopha-ryngeal carcinoma. J transl med 2008; 6: 1.12. lu Cd, Altieri dC: Expression of a novel antia-poptosis gene, survivin correlated with tumor cell apoptosis and p53 accumulation in gastric carcino-mas. Cancer res 1998; 58: 1808-12. 13. Jüttner s, Wissmann C, Jöns t: Vascular En-dothelial Growth Factor-D and its receptor VEGFR-3: two novel independent prognostic mar-kers in gastric adenocarcinoma. J Clin oncol 2006; 24: 228-40. 14. wallner G, ciechański A, dąbrowski A i wsp.: Znaczenie czynników bFGF i VFGF w surowicy krwi u chorych na raka żołądka. Pol Przegl Chir 2002; 74: 1138-44.15. song zJ, Gong P, Wu YE: Relationship betwe-en the expression of iNOS, VEGF, tumor angioge-nesis and gastric cancer. World J Gastroenterol 2002; 8: 591-95.16. Abdou AG, Aiad h: Immunohistochemical eva-luation of vascular endothelial growth factor (VEGF) in colorectal carcinoma. J Egypt natl Canc inst 2006; 18: 311-22.17. Chen X, li Js, liang QP: Expression of PC cell-derived growth factor and vascular endothelial growth factor in esophageal squamous cell carcino-ma and their clinicopathologic significance. Chin med J (Engl) 2008; 121: 881-86.

18. o’Connor ds, schechner Js, Adida C: Control of apoptosis during angiogenesis by survivin expression in endothelial cells. Am J Pathol 2000; 156: 393-98.19. Prewett m, santiago A, oconnor W et al.: Antivascular endothelial growth factor receptor (fetal liver kinase 1) monoclonal antibody inhibits tumor angiogenesis and growth of several mouse and human tumors. Cancer res 1999; 59: 5209-18.20. Yang JC, haworth l, sherry rW et al.: A ran-domized trial of bevacizumab, an antivascular endothelial growth factor antibody, for metastatic renal cancer. n Engl J med 2003; 349: 427-34. 21. Gruchlik A, Chodurek E, dzierewicz z: VEGF-a target of antiangiogenic cancer therapy. Post biol Kom 2007; 34: 557-80.22. Polverini PJ, leibovich sJ: Induction of neova-scularization in vivo and endothelial proliferation in vitro by tumor-associated macrophages. lab invest 1984; 51: 635-42. 23. Xing z, Conway Em, Kang C et al.: Essential role of survivin, an inhibitor of apoptosis protein, in T cell development, maturation, and homeosta-sis. J Exp med 2004; 199: 69-80.24. Plescia j, Salz w, Xia F et al.: Rational design of shepherdin, a novel anticancer agent. Cancer Cell 2005; 5: 457-68.25. Akhtar m, Gallagher m, lisa d: Survivin: role in diagnosis, prognosis, and treatment of bladder cancer. Adv Anat Pathol 2006; 13: 122-26.26. miyachi K, sasaki K, onodera s: Correlation between survivin mRNA expression and lymph node metastasis in gastric cancer. Gastr Cancer 2003; 6: 217-24.27. schultz iJ, Kiemeney lA, Witjes JA: Survivin mRNA expression is elevated in malignant urothe-lial cell carcinomas and predicts time to recurren-ce. Anticancer res 2003; 23: 3327-31.

Received: 24.05.2012 r. Adress correspondence: 20-331 Lublin, ul. Grenadierów 3

Brought to you by | Ryerson UniversityAuthenticated | 141.117.79.62

Download Date | 4/28/13 11:55 AM