- Email: [email protected]
Vascular Endothelial Growth Factor Expression on Prognosis in Vulvar Cancer
GYNECOLOGIC ONCOLOGY ARTICLE NO.
63, 204–209 (1996)
0307
Influence of Microvessel Density and Vascular Permeability Factor/ Vascular Endothelial Growth Factor Expression on Prognosis in Vulvar Cancer ANDREAS OBERMAIR, M.D.,* PETRA KOHLBERGER, M.D.,* DAGMAR BANCHER-TODESCA, M.D.,* CLEMENS TEMPFER, M.D.,* GERHARD SLIUTZ, M.D.,* SEPP LEODOLTER M.D.,* ALEXANDER REINTHALLER, M.D.,*CHRISTIAN KAINZ, M.D.,* GERHARD BREITENECKER, M.D.,† AND GERALD GITSCH, M.D.* *Department of Gynecology and Obstetrics and †Department of Pathology, Gynecopathological Unit, University Hospital of Vienna, Wa¨hringer Gu¨rtel 18-20, A-1090 Vienna, Austria Received February 13, 1996
Microvessel density (MVD) and expression of vascular permeability factor/vascular endothelial growth factor (VPF/VEGF), acting as a highly specific inducer of angiogenesis, were evaluated in tissue specimens of 25 patients with squamous cell cancer of the vulva. MVD was quantified by immunostaining for factor VIIIrelated antigen at one field of 0.25 mm2. VPF/VEGF expression was evaluated immunohistochemically using a monoclonal antiVEGF antibody. FIGO stages I, II, and III were diagnosed in 12, 7, and 6 patients, respectively. MVD ú20/field was found in 10 of 25 tumors and moderate or strong expression of VPF/VEGF in 10 of 25 tumors. High MVD was significantly more frequent in tumors with moderate or strong VPF/VEGF expression compared to tumors with no or weak VPF/VEGF expression (P Å 0.01). Overall survival rates of patients with tumors of high MVD (P Å 0.01) and strong VPF/VEGF expression (P õ 0.01) were significantly poorer compared to those patients with low MVD or poor VPF/VEGF expression. Strong VPF/VEGF expression and high MVD are considered important parameters of tumor angiogenesis and therefore are related to poor survival probability in vulvar cancer patients. q 1996 Academic Press, Inc.
INTRODUCTION
In 1971 Folkman reported the growth of solid tumors and their metastatic spread to be angiogenesis dependent [1]. His findings have been confirmed by several experimental and clinical studies which reported on indirect [2, 3] and direct [4–6] evidence that tumor growth and metastasis are angiogenesis-dependent processes. After switching to the angiogenic phenotype tumors release angiogenic factors, inducing the growth of a capillary net, which surrounds the tumor [7]. Immunohistochemical studies revealed that high microvessel density was associated with poor prognosis in carcinomas
Gyn 4470
/
6d12$$$$$1
MATERIALS AND METHODS
Patients Twenty-five patients with squamous cell carcinoma of the vulva who underwent surgery at the University Hospital of Vienna from 1985 to 1990 were examined retrospectively. Treatment of patients was performed according to Kucera
204
0090-8258/96 $18.00 Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
AID
of the breast [3, 8], the uterine cervix [9, 10], and the ovary [11], and a variety of other malignancies [12]. MVD has been shown to be directly related to the number of circulating tumor cells during surgery for breast cancer [13]. Vascular endothelial growth factor/vascular permeability factor (VEGF/VPF) is a multifunctional cytokine that acts highly specifically as a mitogen on endothelial cells [14, 15]. VEGF/VPF is overexpressed in several malignant tumors, as well as in healing wounds, in rheumatoid arthritis, and in delayed hypersensitivity skin reactions [16]. Four VEGF/ VPF isoforms (polypeptides of 206, 189, 165, and 121 amino acids) have been isolated but apparently express identical biological activities [16]. VEGF has been shown to selectively act on endothelial cells by binding to specific class III receptor tyrosine kinases (flt-1 and KDR) and opening calcium channels to increase intracellular calcium [17]. By these mechanisms VEGF/VPF stimulates angiogenesis by increasing the vascular permeability and by acting as endothelial cell mitogen [18, 19]. This is the first study reporting on patterns of tumor angiogenesis in vulvar cancer. The aim of the present investigation was to describe staining patterns of VPF/VEGF, which is the most specific endothelial mitogen, and microvessel density in vulvar cancer and to detect a possible relationship of these angiogenesis parameters on overall survival.
10-03-96 15:35:04
goas
AP: Gyn Onc
205
ANGIOGENESIS IN VULVAR CANCER
and Weghaupt [20]. Patients with lesions with a depth invasion of 1 mm or less and clinically negative lymph nodes did not receive adjuvant therapy. Patients with a depth of invasion of more than 1 mm and clinically negative lymph nodes underwent adjuvant postoperative groin irradiation. In cases with a strictly lateral location of the tumor (confined to the labium major) only the unilateral groin was irradiated. In cases of medial location or involvement of the labia minora both groins were irradiated. Groin lymph node dissection was performed in cases with clinically suspicious groin lymph nodes. In cases of positive lymph nodes, postoperative irradiation was applied [20]. Histologic staging was performed according to the current International Union against Cancer classification and clinical staging according to the International Federation of Gynecology and Obstetrics (FIGO) classification. Before determination of microvessel density and VEGF expression all cases were reviewed by an experienced pathologist. The investigator of microvessel density (MVD) and VEGF expression was blinded to the clinical outcome of the patients before statistical evaluation.
Staining for VPF/VEGF was performed as follows: As primary antibody, monoclonal anti-human VEGF-neutralizing antibody (R&D Systems Europe, Oxford, UK) was used at a dilution of 1:50 and the sections were incubated 1 hr at room temperature. Sections were then incubated with the detection system (StrAviGen High Performance Immunodetection System; BioGenex). As chromogen, 3-amino-9ethyl-carbozole (BioGenex) was used. Finally sections were washed with distilled water and counterstained with hematoxylin. The staining reaction of VPF/VEGF was confined to the intratumoral area or the tumor margins. Following previous publications intensity of staining for VEGF was graded on a scale of 0–3/, with 0 representing no detectable stain and 3/ representing the strongest stain reaction [22, 23]. Positive controls. The positive control slide was prepared from breast cancer tissue that was known to contain high microvessel density. Negative controls. The negative control slide was prepared from the same tissue block. Instead of using the primary antibody we used a normal, nonimmune serum supernatant.
Immunohistochemistry Surgical specimens were paraffin embedded, stained with hematoxylin and eosin, and reviewed for a sufficient amount of vulvar carcinoma tissue. Two sections, 4 mm each, were then stained immunohistochemically for factor VIII-related antigen (F8-RA) (factor VIII polyclonal; BioGenex, San Ramon, CA), which has been shown to reliably highlight vascular endothelial cells [12, 21]. After dehydration in xylol and graded alcohol, sections were exposed to proteinase for 15 min (protease type XIV; Sigma Chemicals) and 3% H2O2 . Then they were incubated in a humidity chamber at room temperature with the primary antibody (factor VIII polyclonal rabbit; BioGenex) followed by the second antibody (biotinylated anti-rabbit antibody; BioGenex). As chromogen, Fast red (BioGenex) was used. Counterstaining was performed with hematoxylin. After the area of highest neovascularization (hot spot) was located at a total magnification of 1001 by light microscopy, MVD was determined at a total magnification of 2001 using an eyepiece screen with an edge length of 10 mm/100 and an examination area of 0.25 mm2. The staining reaction of F8-RA was strictly confined to the intratumoral area. Counts were done in one field of 0.25 mm2. In the following MVD values are given as the value for one field. Any endothelial cell cluster consisting of two or more cells was considered a single, countable microvessel. Unstained lumina were considered artifacts even if they contained blood or tumor cells. We did not distinguish between blood and lymphatic vessels, since immunostaining for F8-RA does not allow exact differentiation [21].
AID
Gyn 4470
/
6d12$$$$$2
10-03-96 15:35:04
Statistical Methods Univariate analysis of overall survival probability was performed as outlined by Kaplan and Meier [24]. Data on patients who had survived were censored at the last followup visit. By using the nonparametric t test MVD was compared with various clinical and histopathological parameters. MVD was not only analyzed as a continous variable, but also as a dichotomous variable, defining £20 microvessels per field (0.25 mm2) as less vascularized and ú20 per field as highly vascularized. Tumors with no or weak VEGF expression (VEGF grade 0–1) were regarded as VEGF-poor tumors; tumors with moderate (VEGF grade 2) or strong (VEGF grade 3/) VEGF expression were regarded VEGFrich tumors. RESULTS
The mean age of the patient population was 69 years (range 43–89 years). The range of the follow-up was 36– 120 months. At the time of statistical analysis 16 patients were alive and 9 patients had died of disease. The mean overall survival was 49 months (median 36, range 7–72 months). Distribution of FIGO stage, histopathological tumor stage, and histological grade is given in Table 1. Five and 10 years after surgery the probability of overall survival was 33.3 and 22.2%, respectively. Overall mean MVD was 21.3 { 11.5. Mean MVD was 15.8 { 8.5 in patients who survived, and 31.1 { 9.9 in patients who died of disease (t test, P õ 0.01). Type of
goas
AP: Gyn Onc
206
OBERMAIR ET AL.
TABLE 1 Correlation of VEGF Expression (x2 Test) and Microvessel Density (Nonparametric t Test) and Clinical and Histopathological Parameters in 25 Patients with Squamous Cell Cancer of the Vulva No. of cases
No. of VEGF-rich tumors (%)
12 7 6
7 (70.0) 4 (57.1) 2 (40.0)
12 7 6
8 (66.6) 4 (57.1) 2 (40.0)
14 9 2
8 (72.7) 3 (33.3) 2 (100.0)
FIGO stage I II III Tumor stage pT1 pT2 pT3 Grading G1 G2 G3
therapy measures were distributed similarily in the low and the high vascularized group. Fifteen of 25 tumors showed MVD £20/field (low MVD), 10 tumors showed MVD ú20/ field (high MVD). Survival probability was 85.7% after 5 and after 10 years in the low MVD group and 60.0% after 5 years and 37.5% after 10 years in the high MVD group (log-rank P Å 0.01; Fig. 1A). Fifteen of 25 tumors had no or weak VEGF expression (VEGF poor), and 10 tumors showed moderate or strong VEGF expression (VEGF rich). Survival probability was 85.7% after 5 and after 10 years in VEGF-poor tumors and 60.0% after 5 years and 24.0% after 10 years in VEGF-rich
x2
1.94
1.03
4.70
P value
MVD [mean ({SE)]
P value
0.58
24.7 (13.9) 18.0 (11.0) 20.4 (11.2)
0.63
0.59
13.7 (12.9) 18.0 (11.0) 20.4 (11.2)
0.48
0.09
18.7 (12.0) 20.4 (9.7) 27.5 (10.6)
0.34
tumors (log-rank P õ 0.01, Fig. 1B). Immunostaining for VPF/VEGF was found predominantly in the cytoplasm of mast cells, whereas VEGF-staining of endothelial cells or the cytoplasm of tumor cells was detected in two cases only. Positive-stained mast cells were located within the stromal component of the tumor near the tumor border, whereas mast cells at distant sites were negative (Fig. 2). Correlation of VEGF and Vessel Count Intratumoral vessel count was significantly correlated with intense VPF/VEGF expression. Mean MVD was 16.6 { 9.8
FIG. 1. Kaplan–Meier curves for overall survival in (A) patients with less vascularized tumors (MVD £20/field, n Å 15) and with highly vascularized tumors (MVD ú20/field, n Å 10) (log-rank P Å 0.01) and in (B) patients with VEGF grade 0–1 tumors (n Å 15) and with VEGF grade 2–3/ tumors (n Å 10) (log-rank P õ 0.01).
AID
Gyn 4470
/
6d12$$$$$2
10-03-96 15:35:04
goas
AP: Gyn Onc
207
ANGIOGENESIS IN VULVAR CANCER
FIG. 2. (A) Area of highest neovascularization (hot spot) as outlined by high microvessel density (132). (B) Area of strong VEGF expression in mast cells near the tumor border within the stromal component of the tumor (1128).
in VPF/VEGF-poor tumors and 28.4 { 10.7 in VPF/VEGFrich tumors (t test, P Å 0.01). We found high MVD combined in VEGF-rich tumors in 12 cases and low MVD in VEGF-poor tumors in 7 cases. High MVD plus VEGF-poor tumors or low MVD plus VEGF-rich tumors were found in a total of 6 of 25 cases (x2 6.25; P Å 0.01).
AID
Gyn 4470
/
6d12$$$$$2
10-03-96 15:35:04
DISCUSSION
Angiogenesis is essential for tumor growth and metastatic activity and is regulated by a variety of angiogenic molecules. In the present study we report on a significant relationship between morphological findings of tumor angiogenesis,
goas
AP: Gyn Onc
208
OBERMAIR ET AL.
such as MVD and the expression of VPF/VEGF, and the prognosis in vulvar cancer. High MVD was found in 60% of the examined vulvar carcinomas, and moderate or strong VPF/VEGF expression was also found in 60% of the patients. Patients whose tumors revealed high MVD or patients with VEGF-rich tumors showed a significantly poorer overall survival. Tumor angiogenesis is reported to be an independent prognostic factor in various gynecologic [9–11] and nongynecologic tumors [8, 12]. VPF/VEGF is known to be the most specific mitogen for endothelial cells and therefore is considered to play a key role in tumor angiogenesis [25]. However, only few studies report on the association of VEGF expression, expression of VEGF receptors (KDR, flt-1), and microvessel density in human tumors. Toi et al. [22] reported 29 (28.2%) of 109 breast cancer tissues to be VPF/VEGFrich. Takahashi et al. [23] reported not only VPF/VEGF but also VEGF-receptor KDR to be significantly associated with MVD. In both studies VEGF expression was closely associated with high MVD and poor prognosis. MVD was determined by staining for factor VIII-related antigen, and also classification of VPF/VEGF grading was similarily performed in these studies and in the present study. Immunohistochemical studies reported intense and specific staining for VEGF by tumor blood vessels but not by preexisting normal blood vessels [26]. Ultrastructural immunocytochemistry revealed VEGF overexpression of tumor blood vessels at the abluminal plasma membrane and in vesiculovacuolar organelles of tumor-associated microvascular endothelial cells [27]. VEGF overexpression also was found in the cytoplasm of various tumor cells [16], in mononuclear macrophage-like cells [16, 28, 29], and in cancerinfiltrating lymphocytes [30]. In contrast to those previous findings we could not detect clear staining reactions for VEGF in tumor blood vessels and in tumor cells, but we found strong staining reactions for VEGF predominantly in mast cells near the tumor border, whereas mast cells at distant sites were negative. In only 2 of 25 cases we found positive staining reactions for VPF/VEGF in the cytoplasm of tumor cells or in endothelial cells. The implication of these findings will be the subject of further investigation. In normal vulvar tissue a thin layer of microvessels and also singular VPF/VEGF-positive cells can be found at a small zone directly underneath the epidermis (data not shown). Similar to a study on squamous cell cancer of the uterine cervix [10], the area of highest neovascularization was located in the stromal component of the vulvar cancer specimen. As in the paper mentioned above immunostaining for vascular endothelium was performed using factor VIIIrelated antigen because factor VIII-related antigen is considered the golden standard for highlighting vascular endothelial cells. Although CD31 antigen and CD34 antigen represent more sensitive markers, they are less specific [12].
AID
Gyn 4470
/
6d12$$$$$2
10-03-96 15:35:04
Our findings seem to be encouraging but have to be considered preliminary and therefore have to be confirmed by a larger series of patients with multivariate analysis. VPF/ VEGF and MVD seem to be additional parameters for predicting the outcome of patients with vulvar cancer. VPF/ VEGF and its receptors may therefore be considered as a potential target of anti-angiogenic therapeutic strategy. ACKNOWLEDGMENT The present paper was supported by a grant from the ‘‘MedizinischWissenschaftlicher Fonds des Bu¨rgermeisters der Bundeshauptstadt Wien.’’
REFERENCES 1. Folkman, J. Tumor angiogenesis: Therapeutic implications, N. Engl. J. Med. 285, 1182–1186 (1971). 2. Folkman, J. What is the evidence that tumors are angiogenesis dependent? J. Natl. Cancer Inst. 82, 4–6 (1990). 3. Weidner, N., Semple, J. P., Welch, W. R., and Folkman, J. Tumor angiogenesis and metastatis—Correlation in invasive breast carcinoma, N. Engl. J. Med. 324, 1–8 (1991). 4. Ingber, D., Fujita, T., Kishimoto, S., Sudo, K., Kanamaru, T., Brem, H., and Folkman, J. Synthetic analogues of fumagillin that inhibit angiogenesis and suppress tumor growth, Nature 348, 555–557 (1990). 5. Kim, K. J., Li, B., Winer, J., Armani, M., Gillett, N., Phillips, H. S., and Ferrara, N. Inhibition of vascular endothelial growth factor-induced angiogenesis suppresses tumor growth in vivo, Nature 362, 5822–5827 (1993). 6. Millauer, B., Shawyer, L. K., Plate, K. H., Risau, W., and Ullrich, A. Glioblastoma growth inhibited in vivo by a dominant-negative Flk-1 mutant, Nature 364, 576–579 (1994). 7. Folkman, J. Tumor angiogenesis, in Cancer medicine (J. F. Holland, E. Frei III, R. C. Bast Jr., D. W. Kufe, D. L. Morton, and R. R. Weichselbaum, Eds.), Lea & Febiger, Philadelphia, pp. 153–170 (1993). 8. Obermair, A., Kurz, C., Czerwenka, K., Thoma, M., Kaider, A., Wagner, T., Gitsch, G., and Sevelda, P. Microvessel density and vessel invasion in lymph-node-negative breast cancer: Effect on recurrencefree survival, Int. J. Cancer 62, 126–131 (1995). 9. Wiggins, D. L., Granai, C. O., Steinhoff, M. M., and Calabresi, P. Tumor angiogenesis as a prognostic factor in cervical carcinoma, Gynecol. Oncol. 56, 353–356 (1995). 10. Schlenger, K., Ho¨ckel, M., Mitze, M., Scha¨ffer, U., Weikel, W., Knapstein, P. G., and Lambert, A. Tumor vascularity—A novel prognostic factor in advanced cervical carcinoma, Gynecol. Oncol. 59, 57– 66 (1995). 11. Hollingsworth, H. C., Kohn, E. C., Steinberg, S. M., Rothenberg, M. L., and Merino, M. J. Tumor angiogenesis in advanced stage ovarian carcinoma. Am. J. Pathol. 147, 33–41 (1995). 12. Weidner, N. Intratumoral microvessel density as a prognostic factor in cancer, Am. J. Pathol. 147, 9–19 (1995). 13. McCulloch, P., Choy, A., and Martin, L. Association between tumor angiogenesis and tumor cell shedding into effluent venous blood during breast cancer surgery, Lancet 346, 1334–1335 (1995). 14. Leung, D. W., Cachianes, G., Kuang, W., Goeddel, D. V., and Ferrara, N. Vascular endothelial growth factor is a secreted angiogenic mitogen, Science 246, 1306–1309 (1989).
goas
AP: Gyn Onc
ANGIOGENESIS IN VULVAR CANCER 15. Senger, D. R., Conolly, D. T., Van de Water, L., Feder, J., and Dvorak, H. F. Purification and NH2-terminal amino acid sequence of guinea pig tumor-secreted vascular permeability factor, Cancer Res. 50, 1774– 1778 (1990). 16. Dvorak, H. F., Brown, L. F., Detmar, M., and Dvorak, A. M. Vascular permeability factor/vascular endothelial growth factor, microvascular hyperpermeability, and angiogenesis, Am. J. Pathol. 146, 1029–1039 (1995). 17. Brock, T. A., Dvorak, H. F., and Senger, D. R. Tumor-secreted vascular permeability factor increases cytosolic Ca2/ and von Willebrand factor release in human endothelial cells, Am. J. Pathol. 138, 213–221 (1991).
24. 25.
26.
18. Ferrara, N., Heinsohn, H., Walder, C. E., Bunting, S., and Thomas, G. R. The regulation of blood vessel growth by vascular endothelial growth factor, Ann. N. Y. Acad. Sci. 752, 246–256 (1995). 19. Millauer, B., Wizigman-Voos, S., Schnurch, H., Martinez, R., Moller, N. P., Risau, W., and Ullrich, A. High affinity VEGF binding and developmental expression suggest Flk-1 as a major regulator of vasculogenesis and angiogenesis, Cell 72, 835–846 (1993). 20. Kucera, H., and Weghaupt, K. The electrosurgical operation of vulvar carcinoma with postoperative irradiation of inguinal lymph nodes, Gynecol. Oncol. 29, 158–167 (1988).
27.
28.
21. Lee, A. K. Basement membrane and endothelial agents: Their role in evaluation of tumor invasion and metastasis, in Advances in immunohistochemistry (R. A. De Ellis, Ed.), Raven Press, New York, pp. 363– 393 (1988).
29.
22. Toi, M., Hoshina, S., Takayanagi, T., and Tominaga, T. Association of vascular endothelial growth factor expression with tumor angiogenesis and with early relapse in primary breast cancer, Jpn. J. Cancer Res. 85, 1045–1049 (1994).
30.
23. Takahashi, Y., Kitadai, Y., Bucana, C. D., Cleary, K. R., and Ellis, L. M. Expression of vascular endothelial growth factor and its receptor,
AID
Gyn 4470
/
6d12$$$$$3
10-03-96 15:35:04
209
KDR, correlates with vascularity metastasis, and proliferation of human colon cancer, Cancer Res. 55, 3964–3968 (1995). Kaplan, E. L., and Meier, P. Non-parametric estimation from incomplete observations, J. Am. Stat. Assoc. 53, 457–481 (1985). Zhang, H. T., Craft, P., Scott, P. A., Ziche, M., Weich, H. A., Harris, A. L., and Bicknell, R. Enhancement of tumor growth and vascular density by transfection of vascular endothelial cell growth factor into MCF-7 human breast carcinoma cells, J. Natl. Cancer Inst. 87, 213– 219 (1995). Dvorak, H. F., Sioussat, T. M., Brown, L. F., Berse, B., Nagy, J., Sotrel, A., Manseau, E. J., Van De Water, L., and Senger, D. R. Distribution of vascular permeability factor (vascular endothelial growth factor) in tumors: Concentration in tumor blood vessels, J. Exp. Med. 174, 1275– 1278 (1991). Qu-Hong, Nagy, J. A., Senger, D. R., Dvorak, H. F., and Dvorak, A. M. Ultrastructural localization of vascular permeability factor/vascular endothelial growth factor (VPF/VEGF) to the abluminal plasma membrane and vesiculovacuolar organelles of tumor microvascular endothelium, J. Histochem. Cytochem. 43, 381–389 (1995). Brown, L. F., Yeo, K. T., Berse, B., Yeo, T. K., Senger, D. R., Dvorak, H. F., and Van De Water, L. Expression of vascular permeability factor (vascular endothelial growth factor) by epidermal keratinocytes during wound healing, J. Exp. Med. 176, 1375–1379 (1992). Hashimoto, M., Ohsawa, M., Ohnishi, A., Naka, N., Hirota, S., Kitamura, Y., and Aozasa, K. Expression of vascular endothelial growth factor and its receptor mRNA in angiosarcoma. Lab. Invest. 73, 859– 863 (1995). Freeman, M. R., Schneck, F. X., Gagnon, M. L., Corless, C., Soker, S., Niknejad, K., Peoples, G. E., and Klagsbrun, M. Peripheral blood T lymphocytes and lymphocytes infiltrating human cancers express vascular endothelial growth factor: A potential role for T cells in angiogenesis. Cancer Res. 55, 4140–4145 (1995).
goas
AP: Gyn Onc
63, 204–209 (1996)
0307
Influence of Microvessel Density and Vascular Permeability Factor/ Vascular Endothelial Growth Factor Expression on Prognosis in Vulvar Cancer ANDREAS OBERMAIR, M.D.,* PETRA KOHLBERGER, M.D.,* DAGMAR BANCHER-TODESCA, M.D.,* CLEMENS TEMPFER, M.D.,* GERHARD SLIUTZ, M.D.,* SEPP LEODOLTER M.D.,* ALEXANDER REINTHALLER, M.D.,*CHRISTIAN KAINZ, M.D.,* GERHARD BREITENECKER, M.D.,† AND GERALD GITSCH, M.D.* *Department of Gynecology and Obstetrics and †Department of Pathology, Gynecopathological Unit, University Hospital of Vienna, Wa¨hringer Gu¨rtel 18-20, A-1090 Vienna, Austria Received February 13, 1996
Microvessel density (MVD) and expression of vascular permeability factor/vascular endothelial growth factor (VPF/VEGF), acting as a highly specific inducer of angiogenesis, were evaluated in tissue specimens of 25 patients with squamous cell cancer of the vulva. MVD was quantified by immunostaining for factor VIIIrelated antigen at one field of 0.25 mm2. VPF/VEGF expression was evaluated immunohistochemically using a monoclonal antiVEGF antibody. FIGO stages I, II, and III were diagnosed in 12, 7, and 6 patients, respectively. MVD ú20/field was found in 10 of 25 tumors and moderate or strong expression of VPF/VEGF in 10 of 25 tumors. High MVD was significantly more frequent in tumors with moderate or strong VPF/VEGF expression compared to tumors with no or weak VPF/VEGF expression (P Å 0.01). Overall survival rates of patients with tumors of high MVD (P Å 0.01) and strong VPF/VEGF expression (P õ 0.01) were significantly poorer compared to those patients with low MVD or poor VPF/VEGF expression. Strong VPF/VEGF expression and high MVD are considered important parameters of tumor angiogenesis and therefore are related to poor survival probability in vulvar cancer patients. q 1996 Academic Press, Inc.
INTRODUCTION
In 1971 Folkman reported the growth of solid tumors and their metastatic spread to be angiogenesis dependent [1]. His findings have been confirmed by several experimental and clinical studies which reported on indirect [2, 3] and direct [4–6] evidence that tumor growth and metastasis are angiogenesis-dependent processes. After switching to the angiogenic phenotype tumors release angiogenic factors, inducing the growth of a capillary net, which surrounds the tumor [7]. Immunohistochemical studies revealed that high microvessel density was associated with poor prognosis in carcinomas
Gyn 4470
/
6d12$$$$$1
MATERIALS AND METHODS
Patients Twenty-five patients with squamous cell carcinoma of the vulva who underwent surgery at the University Hospital of Vienna from 1985 to 1990 were examined retrospectively. Treatment of patients was performed according to Kucera
204
0090-8258/96 $18.00 Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
AID
of the breast [3, 8], the uterine cervix [9, 10], and the ovary [11], and a variety of other malignancies [12]. MVD has been shown to be directly related to the number of circulating tumor cells during surgery for breast cancer [13]. Vascular endothelial growth factor/vascular permeability factor (VEGF/VPF) is a multifunctional cytokine that acts highly specifically as a mitogen on endothelial cells [14, 15]. VEGF/VPF is overexpressed in several malignant tumors, as well as in healing wounds, in rheumatoid arthritis, and in delayed hypersensitivity skin reactions [16]. Four VEGF/ VPF isoforms (polypeptides of 206, 189, 165, and 121 amino acids) have been isolated but apparently express identical biological activities [16]. VEGF has been shown to selectively act on endothelial cells by binding to specific class III receptor tyrosine kinases (flt-1 and KDR) and opening calcium channels to increase intracellular calcium [17]. By these mechanisms VEGF/VPF stimulates angiogenesis by increasing the vascular permeability and by acting as endothelial cell mitogen [18, 19]. This is the first study reporting on patterns of tumor angiogenesis in vulvar cancer. The aim of the present investigation was to describe staining patterns of VPF/VEGF, which is the most specific endothelial mitogen, and microvessel density in vulvar cancer and to detect a possible relationship of these angiogenesis parameters on overall survival.
10-03-96 15:35:04
goas
AP: Gyn Onc
205
ANGIOGENESIS IN VULVAR CANCER
and Weghaupt [20]. Patients with lesions with a depth invasion of 1 mm or less and clinically negative lymph nodes did not receive adjuvant therapy. Patients with a depth of invasion of more than 1 mm and clinically negative lymph nodes underwent adjuvant postoperative groin irradiation. In cases with a strictly lateral location of the tumor (confined to the labium major) only the unilateral groin was irradiated. In cases of medial location or involvement of the labia minora both groins were irradiated. Groin lymph node dissection was performed in cases with clinically suspicious groin lymph nodes. In cases of positive lymph nodes, postoperative irradiation was applied [20]. Histologic staging was performed according to the current International Union against Cancer classification and clinical staging according to the International Federation of Gynecology and Obstetrics (FIGO) classification. Before determination of microvessel density and VEGF expression all cases were reviewed by an experienced pathologist. The investigator of microvessel density (MVD) and VEGF expression was blinded to the clinical outcome of the patients before statistical evaluation.
Staining for VPF/VEGF was performed as follows: As primary antibody, monoclonal anti-human VEGF-neutralizing antibody (R&D Systems Europe, Oxford, UK) was used at a dilution of 1:50 and the sections were incubated 1 hr at room temperature. Sections were then incubated with the detection system (StrAviGen High Performance Immunodetection System; BioGenex). As chromogen, 3-amino-9ethyl-carbozole (BioGenex) was used. Finally sections were washed with distilled water and counterstained with hematoxylin. The staining reaction of VPF/VEGF was confined to the intratumoral area or the tumor margins. Following previous publications intensity of staining for VEGF was graded on a scale of 0–3/, with 0 representing no detectable stain and 3/ representing the strongest stain reaction [22, 23]. Positive controls. The positive control slide was prepared from breast cancer tissue that was known to contain high microvessel density. Negative controls. The negative control slide was prepared from the same tissue block. Instead of using the primary antibody we used a normal, nonimmune serum supernatant.
Immunohistochemistry Surgical specimens were paraffin embedded, stained with hematoxylin and eosin, and reviewed for a sufficient amount of vulvar carcinoma tissue. Two sections, 4 mm each, were then stained immunohistochemically for factor VIII-related antigen (F8-RA) (factor VIII polyclonal; BioGenex, San Ramon, CA), which has been shown to reliably highlight vascular endothelial cells [12, 21]. After dehydration in xylol and graded alcohol, sections were exposed to proteinase for 15 min (protease type XIV; Sigma Chemicals) and 3% H2O2 . Then they were incubated in a humidity chamber at room temperature with the primary antibody (factor VIII polyclonal rabbit; BioGenex) followed by the second antibody (biotinylated anti-rabbit antibody; BioGenex). As chromogen, Fast red (BioGenex) was used. Counterstaining was performed with hematoxylin. After the area of highest neovascularization (hot spot) was located at a total magnification of 1001 by light microscopy, MVD was determined at a total magnification of 2001 using an eyepiece screen with an edge length of 10 mm/100 and an examination area of 0.25 mm2. The staining reaction of F8-RA was strictly confined to the intratumoral area. Counts were done in one field of 0.25 mm2. In the following MVD values are given as the value for one field. Any endothelial cell cluster consisting of two or more cells was considered a single, countable microvessel. Unstained lumina were considered artifacts even if they contained blood or tumor cells. We did not distinguish between blood and lymphatic vessels, since immunostaining for F8-RA does not allow exact differentiation [21].
AID
Gyn 4470
/
6d12$$$$$2
10-03-96 15:35:04
Statistical Methods Univariate analysis of overall survival probability was performed as outlined by Kaplan and Meier [24]. Data on patients who had survived were censored at the last followup visit. By using the nonparametric t test MVD was compared with various clinical and histopathological parameters. MVD was not only analyzed as a continous variable, but also as a dichotomous variable, defining £20 microvessels per field (0.25 mm2) as less vascularized and ú20 per field as highly vascularized. Tumors with no or weak VEGF expression (VEGF grade 0–1) were regarded as VEGF-poor tumors; tumors with moderate (VEGF grade 2) or strong (VEGF grade 3/) VEGF expression were regarded VEGFrich tumors. RESULTS
The mean age of the patient population was 69 years (range 43–89 years). The range of the follow-up was 36– 120 months. At the time of statistical analysis 16 patients were alive and 9 patients had died of disease. The mean overall survival was 49 months (median 36, range 7–72 months). Distribution of FIGO stage, histopathological tumor stage, and histological grade is given in Table 1. Five and 10 years after surgery the probability of overall survival was 33.3 and 22.2%, respectively. Overall mean MVD was 21.3 { 11.5. Mean MVD was 15.8 { 8.5 in patients who survived, and 31.1 { 9.9 in patients who died of disease (t test, P õ 0.01). Type of
goas
AP: Gyn Onc
206
OBERMAIR ET AL.
TABLE 1 Correlation of VEGF Expression (x2 Test) and Microvessel Density (Nonparametric t Test) and Clinical and Histopathological Parameters in 25 Patients with Squamous Cell Cancer of the Vulva No. of cases
No. of VEGF-rich tumors (%)
12 7 6
7 (70.0) 4 (57.1) 2 (40.0)
12 7 6
8 (66.6) 4 (57.1) 2 (40.0)
14 9 2
8 (72.7) 3 (33.3) 2 (100.0)
FIGO stage I II III Tumor stage pT1 pT2 pT3 Grading G1 G2 G3
therapy measures were distributed similarily in the low and the high vascularized group. Fifteen of 25 tumors showed MVD £20/field (low MVD), 10 tumors showed MVD ú20/ field (high MVD). Survival probability was 85.7% after 5 and after 10 years in the low MVD group and 60.0% after 5 years and 37.5% after 10 years in the high MVD group (log-rank P Å 0.01; Fig. 1A). Fifteen of 25 tumors had no or weak VEGF expression (VEGF poor), and 10 tumors showed moderate or strong VEGF expression (VEGF rich). Survival probability was 85.7% after 5 and after 10 years in VEGF-poor tumors and 60.0% after 5 years and 24.0% after 10 years in VEGF-rich
x2
1.94
1.03
4.70
P value
MVD [mean ({SE)]
P value
0.58
24.7 (13.9) 18.0 (11.0) 20.4 (11.2)
0.63
0.59
13.7 (12.9) 18.0 (11.0) 20.4 (11.2)
0.48
0.09
18.7 (12.0) 20.4 (9.7) 27.5 (10.6)
0.34
tumors (log-rank P õ 0.01, Fig. 1B). Immunostaining for VPF/VEGF was found predominantly in the cytoplasm of mast cells, whereas VEGF-staining of endothelial cells or the cytoplasm of tumor cells was detected in two cases only. Positive-stained mast cells were located within the stromal component of the tumor near the tumor border, whereas mast cells at distant sites were negative (Fig. 2). Correlation of VEGF and Vessel Count Intratumoral vessel count was significantly correlated with intense VPF/VEGF expression. Mean MVD was 16.6 { 9.8
FIG. 1. Kaplan–Meier curves for overall survival in (A) patients with less vascularized tumors (MVD £20/field, n Å 15) and with highly vascularized tumors (MVD ú20/field, n Å 10) (log-rank P Å 0.01) and in (B) patients with VEGF grade 0–1 tumors (n Å 15) and with VEGF grade 2–3/ tumors (n Å 10) (log-rank P õ 0.01).
AID
Gyn 4470
/
6d12$$$$$2
10-03-96 15:35:04
goas
AP: Gyn Onc
207
ANGIOGENESIS IN VULVAR CANCER
FIG. 2. (A) Area of highest neovascularization (hot spot) as outlined by high microvessel density (132). (B) Area of strong VEGF expression in mast cells near the tumor border within the stromal component of the tumor (1128).
in VPF/VEGF-poor tumors and 28.4 { 10.7 in VPF/VEGFrich tumors (t test, P Å 0.01). We found high MVD combined in VEGF-rich tumors in 12 cases and low MVD in VEGF-poor tumors in 7 cases. High MVD plus VEGF-poor tumors or low MVD plus VEGF-rich tumors were found in a total of 6 of 25 cases (x2 6.25; P Å 0.01).
AID
Gyn 4470
/
6d12$$$$$2
10-03-96 15:35:04
DISCUSSION
Angiogenesis is essential for tumor growth and metastatic activity and is regulated by a variety of angiogenic molecules. In the present study we report on a significant relationship between morphological findings of tumor angiogenesis,
goas
AP: Gyn Onc
208
OBERMAIR ET AL.
such as MVD and the expression of VPF/VEGF, and the prognosis in vulvar cancer. High MVD was found in 60% of the examined vulvar carcinomas, and moderate or strong VPF/VEGF expression was also found in 60% of the patients. Patients whose tumors revealed high MVD or patients with VEGF-rich tumors showed a significantly poorer overall survival. Tumor angiogenesis is reported to be an independent prognostic factor in various gynecologic [9–11] and nongynecologic tumors [8, 12]. VPF/VEGF is known to be the most specific mitogen for endothelial cells and therefore is considered to play a key role in tumor angiogenesis [25]. However, only few studies report on the association of VEGF expression, expression of VEGF receptors (KDR, flt-1), and microvessel density in human tumors. Toi et al. [22] reported 29 (28.2%) of 109 breast cancer tissues to be VPF/VEGFrich. Takahashi et al. [23] reported not only VPF/VEGF but also VEGF-receptor KDR to be significantly associated with MVD. In both studies VEGF expression was closely associated with high MVD and poor prognosis. MVD was determined by staining for factor VIII-related antigen, and also classification of VPF/VEGF grading was similarily performed in these studies and in the present study. Immunohistochemical studies reported intense and specific staining for VEGF by tumor blood vessels but not by preexisting normal blood vessels [26]. Ultrastructural immunocytochemistry revealed VEGF overexpression of tumor blood vessels at the abluminal plasma membrane and in vesiculovacuolar organelles of tumor-associated microvascular endothelial cells [27]. VEGF overexpression also was found in the cytoplasm of various tumor cells [16], in mononuclear macrophage-like cells [16, 28, 29], and in cancerinfiltrating lymphocytes [30]. In contrast to those previous findings we could not detect clear staining reactions for VEGF in tumor blood vessels and in tumor cells, but we found strong staining reactions for VEGF predominantly in mast cells near the tumor border, whereas mast cells at distant sites were negative. In only 2 of 25 cases we found positive staining reactions for VPF/VEGF in the cytoplasm of tumor cells or in endothelial cells. The implication of these findings will be the subject of further investigation. In normal vulvar tissue a thin layer of microvessels and also singular VPF/VEGF-positive cells can be found at a small zone directly underneath the epidermis (data not shown). Similar to a study on squamous cell cancer of the uterine cervix [10], the area of highest neovascularization was located in the stromal component of the vulvar cancer specimen. As in the paper mentioned above immunostaining for vascular endothelium was performed using factor VIIIrelated antigen because factor VIII-related antigen is considered the golden standard for highlighting vascular endothelial cells. Although CD31 antigen and CD34 antigen represent more sensitive markers, they are less specific [12].
AID
Gyn 4470
/
6d12$$$$$2
10-03-96 15:35:04
Our findings seem to be encouraging but have to be considered preliminary and therefore have to be confirmed by a larger series of patients with multivariate analysis. VPF/ VEGF and MVD seem to be additional parameters for predicting the outcome of patients with vulvar cancer. VPF/ VEGF and its receptors may therefore be considered as a potential target of anti-angiogenic therapeutic strategy. ACKNOWLEDGMENT The present paper was supported by a grant from the ‘‘MedizinischWissenschaftlicher Fonds des Bu¨rgermeisters der Bundeshauptstadt Wien.’’
REFERENCES 1. Folkman, J. Tumor angiogenesis: Therapeutic implications, N. Engl. J. Med. 285, 1182–1186 (1971). 2. Folkman, J. What is the evidence that tumors are angiogenesis dependent? J. Natl. Cancer Inst. 82, 4–6 (1990). 3. Weidner, N., Semple, J. P., Welch, W. R., and Folkman, J. Tumor angiogenesis and metastatis—Correlation in invasive breast carcinoma, N. Engl. J. Med. 324, 1–8 (1991). 4. Ingber, D., Fujita, T., Kishimoto, S., Sudo, K., Kanamaru, T., Brem, H., and Folkman, J. Synthetic analogues of fumagillin that inhibit angiogenesis and suppress tumor growth, Nature 348, 555–557 (1990). 5. Kim, K. J., Li, B., Winer, J., Armani, M., Gillett, N., Phillips, H. S., and Ferrara, N. Inhibition of vascular endothelial growth factor-induced angiogenesis suppresses tumor growth in vivo, Nature 362, 5822–5827 (1993). 6. Millauer, B., Shawyer, L. K., Plate, K. H., Risau, W., and Ullrich, A. Glioblastoma growth inhibited in vivo by a dominant-negative Flk-1 mutant, Nature 364, 576–579 (1994). 7. Folkman, J. Tumor angiogenesis, in Cancer medicine (J. F. Holland, E. Frei III, R. C. Bast Jr., D. W. Kufe, D. L. Morton, and R. R. Weichselbaum, Eds.), Lea & Febiger, Philadelphia, pp. 153–170 (1993). 8. Obermair, A., Kurz, C., Czerwenka, K., Thoma, M., Kaider, A., Wagner, T., Gitsch, G., and Sevelda, P. Microvessel density and vessel invasion in lymph-node-negative breast cancer: Effect on recurrencefree survival, Int. J. Cancer 62, 126–131 (1995). 9. Wiggins, D. L., Granai, C. O., Steinhoff, M. M., and Calabresi, P. Tumor angiogenesis as a prognostic factor in cervical carcinoma, Gynecol. Oncol. 56, 353–356 (1995). 10. Schlenger, K., Ho¨ckel, M., Mitze, M., Scha¨ffer, U., Weikel, W., Knapstein, P. G., and Lambert, A. Tumor vascularity—A novel prognostic factor in advanced cervical carcinoma, Gynecol. Oncol. 59, 57– 66 (1995). 11. Hollingsworth, H. C., Kohn, E. C., Steinberg, S. M., Rothenberg, M. L., and Merino, M. J. Tumor angiogenesis in advanced stage ovarian carcinoma. Am. J. Pathol. 147, 33–41 (1995). 12. Weidner, N. Intratumoral microvessel density as a prognostic factor in cancer, Am. J. Pathol. 147, 9–19 (1995). 13. McCulloch, P., Choy, A., and Martin, L. Association between tumor angiogenesis and tumor cell shedding into effluent venous blood during breast cancer surgery, Lancet 346, 1334–1335 (1995). 14. Leung, D. W., Cachianes, G., Kuang, W., Goeddel, D. V., and Ferrara, N. Vascular endothelial growth factor is a secreted angiogenic mitogen, Science 246, 1306–1309 (1989).
goas
AP: Gyn Onc
ANGIOGENESIS IN VULVAR CANCER 15. Senger, D. R., Conolly, D. T., Van de Water, L., Feder, J., and Dvorak, H. F. Purification and NH2-terminal amino acid sequence of guinea pig tumor-secreted vascular permeability factor, Cancer Res. 50, 1774– 1778 (1990). 16. Dvorak, H. F., Brown, L. F., Detmar, M., and Dvorak, A. M. Vascular permeability factor/vascular endothelial growth factor, microvascular hyperpermeability, and angiogenesis, Am. J. Pathol. 146, 1029–1039 (1995). 17. Brock, T. A., Dvorak, H. F., and Senger, D. R. Tumor-secreted vascular permeability factor increases cytosolic Ca2/ and von Willebrand factor release in human endothelial cells, Am. J. Pathol. 138, 213–221 (1991).
24. 25.
26.
18. Ferrara, N., Heinsohn, H., Walder, C. E., Bunting, S., and Thomas, G. R. The regulation of blood vessel growth by vascular endothelial growth factor, Ann. N. Y. Acad. Sci. 752, 246–256 (1995). 19. Millauer, B., Wizigman-Voos, S., Schnurch, H., Martinez, R., Moller, N. P., Risau, W., and Ullrich, A. High affinity VEGF binding and developmental expression suggest Flk-1 as a major regulator of vasculogenesis and angiogenesis, Cell 72, 835–846 (1993). 20. Kucera, H., and Weghaupt, K. The electrosurgical operation of vulvar carcinoma with postoperative irradiation of inguinal lymph nodes, Gynecol. Oncol. 29, 158–167 (1988).
27.
28.
21. Lee, A. K. Basement membrane and endothelial agents: Their role in evaluation of tumor invasion and metastasis, in Advances in immunohistochemistry (R. A. De Ellis, Ed.), Raven Press, New York, pp. 363– 393 (1988).
29.
22. Toi, M., Hoshina, S., Takayanagi, T., and Tominaga, T. Association of vascular endothelial growth factor expression with tumor angiogenesis and with early relapse in primary breast cancer, Jpn. J. Cancer Res. 85, 1045–1049 (1994).
30.
23. Takahashi, Y., Kitadai, Y., Bucana, C. D., Cleary, K. R., and Ellis, L. M. Expression of vascular endothelial growth factor and its receptor,
AID
Gyn 4470
/
6d12$$$$$3
10-03-96 15:35:04
209
KDR, correlates with vascularity metastasis, and proliferation of human colon cancer, Cancer Res. 55, 3964–3968 (1995). Kaplan, E. L., and Meier, P. Non-parametric estimation from incomplete observations, J. Am. Stat. Assoc. 53, 457–481 (1985). Zhang, H. T., Craft, P., Scott, P. A., Ziche, M., Weich, H. A., Harris, A. L., and Bicknell, R. Enhancement of tumor growth and vascular density by transfection of vascular endothelial cell growth factor into MCF-7 human breast carcinoma cells, J. Natl. Cancer Inst. 87, 213– 219 (1995). Dvorak, H. F., Sioussat, T. M., Brown, L. F., Berse, B., Nagy, J., Sotrel, A., Manseau, E. J., Van De Water, L., and Senger, D. R. Distribution of vascular permeability factor (vascular endothelial growth factor) in tumors: Concentration in tumor blood vessels, J. Exp. Med. 174, 1275– 1278 (1991). Qu-Hong, Nagy, J. A., Senger, D. R., Dvorak, H. F., and Dvorak, A. M. Ultrastructural localization of vascular permeability factor/vascular endothelial growth factor (VPF/VEGF) to the abluminal plasma membrane and vesiculovacuolar organelles of tumor microvascular endothelium, J. Histochem. Cytochem. 43, 381–389 (1995). Brown, L. F., Yeo, K. T., Berse, B., Yeo, T. K., Senger, D. R., Dvorak, H. F., and Van De Water, L. Expression of vascular permeability factor (vascular endothelial growth factor) by epidermal keratinocytes during wound healing, J. Exp. Med. 176, 1375–1379 (1992). Hashimoto, M., Ohsawa, M., Ohnishi, A., Naka, N., Hirota, S., Kitamura, Y., and Aozasa, K. Expression of vascular endothelial growth factor and its receptor mRNA in angiosarcoma. Lab. Invest. 73, 859– 863 (1995). Freeman, M. R., Schneck, F. X., Gagnon, M. L., Corless, C., Soker, S., Niknejad, K., Peoples, G. E., and Klagsbrun, M. Peripheral blood T lymphocytes and lymphocytes infiltrating human cancers express vascular endothelial growth factor: A potential role for T cells in angiogenesis. Cancer Res. 55, 4140–4145 (1995).
goas
AP: Gyn Onc