Prognostic Significance of Vascular Endothelial Growth Factor D in Gastric Carcinoma

Abstract

The angiogenic factor called vascular endothelial growth factor (VEGF)-D is a ligand for VEGF receptor-2 (VEGFR-2/KDR) and receptor-3 (VEGFR-3/Flt-4). It is implicated in the development of lymphatic vessels and promotion of lymphatic metastasis. The purpose of this study was to investigate the prognostic significance of VEGF-D expression in patients with gastric carcinoma. We assessed the expression of VEGF-D in gastric carcinoma by immunohistochemistry on 143 consecutive patients' stored sections and evaluated the lymphatic vessel count (LVC) in tumors using the novel selective lymphatic endothelium marker D2-40. VEGF-D expression was observed in 55 (39%) tumor sections. The expression of VEGF-D correlated significantly with tumor size, T of the TNM classification, lymphatic and venous system invasion, LVC, lymph node metastasis, M of TNM, and pTNM stage. Multivariate analysis indicated that VEGF-D expression was an independent prognostic factor for both relapse-free survival (RFS) and overall survival (OS). Our data indicate the involvement of VEGF-D in tumor progression via lymphoangiogenic pathways. Practically, VEGF-D expression can be useful for predicting RFS and OS in patients with gastric carcinoma.

Full text

Prognostic Significance of Vascular Endothelial
Growth Factor D in Gastric Carcinoma
Atsuo Shida, MD,
1
Shuichi Fujioka, MD,
1
Yoshio Ishibashi, MD,
1
Katsutoshi Kobayashi, MD,
1
Hiroshi Nimura, MD,
1
Norio Mitsumori, MD,
1
Yutaka Suzuki, MD,
1
Makio Kawakami, MD,
2
Mitsuyoshi Urashima, MD,
3
Katsuhiko Yanaga, MD
1
1
Department of Surgery, Jikei University School of Medicine, Nishi-shinbashi, 3-25-8 Minato-ku, Tokyo 105-8461, Japan
2
Department of Pathology, Jikei University School of Medicine, Nishi-shinbashi, 3-25-8 Minato-ku, Tokyo 105-8461,
Japan
3
Division of Clinical Research and Development, Jikei University School of Medicine, Nishi-shinbashi, 3-25-8
Minato-ku, Tokyo 105-8461, Japan
Abstract
The angiogenic factor called vascular endothelial growth factor (VEGF)-D is a ligand for VEGF
receptor-2 (VEGFR-2/KDR) and receptor-3 (VEGFR-3/Flt-4). It is implicated in the development of
lymphatic vessels and promotion of lymphatic metastasis. The purpose of this study was to
investigate the prognostic significance of VEGF-D expression in patients with gastric carcinoma.
We assessed the expression of VEGF-D in gastric carcinoma by immunohistochemistry on 143
consecutive patients’ stored sections and evaluated the lymphatic vessel count (LVC) in tumors
using the novel selective lymphatic endothelium marker D2-40. VEGF-D expression was observed
in 55 (39%) tumor sections. The expression of VEGF-D correlated significantly with tumor size, T
of the TNM classification, lymphatic and venous system invasion, LVC, lymph node metastasis, M
of TNM, and pTNM stage. Multivariate analysis indicated that VEGF-D expression was an inde-
pendent prognostic factor for both relapse-free survival (RFS) and overall survival (OS). Our data
indicate the involvement of VEGF-D in tumor progression via lymphoangiogenic pathways.
Practically, VEGF-D expression can be useful for predicting RFS and OS in patients with gastric
carcinoma.
G
astric cancer is one of the leading causes of cancer
deaths worldwide.
1
The extent of regional lymph
node metastasis is an important indicator of tumor
aggressiveness and forms the basis for surgical ablation
of the local lymph nodes.
2
Lymphatic capillaries are
thought to provide entrance into the lymphatic vascular
system when the tumor cells migrate to the lymph node.
3
Vascular endothelial growth factor (VEGF), a member
of the platelet-derived growth factor family, is a major
inducer of angiogenesis and vessel permeability.
4,5
Five
additional members of this family—VEGF-B, VEGF-C,
VEGF-D, VEGF-E, and placenta growth factor
(PlGF)—have been characterized.
6
VEGF-C and VEGF-
D are ligands for VEGFR-3 (Flt-4), a tyrosine kinase
receptor expressed predominantly in lymphatic endothe-
lial cells.
7
VEGF-C and VEGF-D also bind to VEGFR-2
(KDR/FLK-1), which is mainly expressed in vascular
endothelial cells and is supposed to take part in lym-
phangiogenesis and angiogenesis. Although the relation
Correspondence to: Shuichi Fujioka, MD, e-mail: sfujioka@jikei.ac.
jp
 2005 by the Socie
´
te
´
Internationale de Chirurgie World J Surg (2005) 29: 1–8
Published Online: 16 November 2005 DOI: 10.1007/s00268-005-0076-z

between VEGF-D expression and lymph node metastasis
has been reported in gastric carcinoma,
8,9
there are no
published articles regarding the impact of VEGF-D
expression of gastric carcinoma specimens on patient
outcome. Therefore we investigated the prognostic sig-
nificance of the expression of VEGF-D in patients with
gastric carcinoma.
MATERIALS AND METHODS
Patients and Tumor Samples
Paraffin-embedded stored specimens from 143 con-
secutive patients with primary gastric adenocarcinoma
diagnosed and treated by gastrectomy with standard
lymph node dissection at the Second Department of
Surgery, Jikei University Hospital between January 1997
and December 1998 were studied using immunohisto-
chemistry. Tumor stages and the definition of criteria for
histologic classification followed those proposed by the
Japanese Society for the Research of Gastric Cancer
(JRSGC).
10
Histologic grade was classified into two
groups based on the predominant features: The differ-
entiated group contained patients with well differentiated
and moderately differentiated adenocarcinoma and pap-
illary adenocarcinoma; and the undifferentiated group
contained patients with poorly differentiated carcinoma,
signet-ring cell carcinoma, and mucinous adenocarci-
noma.
Immunohistochemical Staining for VEGF-D and
Lymphatic Endothelium
Using paraffin-embedded specimens from patients
with gastric cancer, VEGF-D protein was detected using
the anti-VEGF-D goat polyclonal antibody (R&D System,
Minneapolis, MN, USA) at 10 lg/ml concentration fol-
lowed by the immunoperoxidase method using a Histo-
fine SAB-PO(M) kit (Nichirei, Tokyo, Japan). Using slices
from the same paraffin-embedded specimens, lymphatic
endothelium was stained with anti-human endothelium
murine monoclonal antibody at 1:100 dilution for D2-40
(Signet Laboratories, Dedham, MA, USA)
11
followed by
a Omunitag/HRP kit (Thermo Electron, Waltham, MA,
USA). Control goat and murine immunoglobulin G were
used for each staining, respectively (Daiichi Fine
Chemical, Takaoka, Toyama, Japan).
The examiners were kept unaware of the patients’
clinical and histologic (hematoxylin-eosin staining) infor-
mation when judging that the staining was positive or
negative. Two investigators (A.S. and Y.I.) evaluated the
staining levels independently, after which discordant
evaluations were adjusted through the connected micro-
scope. The intensity of VEGF-D staining was graded on a
scale of 0 to 3+ under a ·200 field: 0, no detectable
cytoplasmic staining of cancer cells; 1+, weak staining;
2+, moderate staining; 3+, strong staining. A specimen
was regarded as positive when the intensity of staining
was 2+ or 3+ and negative when the intensity of staining
was 0 or 1+.
The lymphatic vessel count (LVC) was assessed by
light microscopic examination of the tumor region con-
taining the largest number of capillaries and small ven-
ules. At low magnifications (·40 and ·100), highly
vascular areas were identified by scanning tumor sec-
tions. After the area with the most neovascularization was
identified, lymphatic vessels were counted from three
high power fields (HPF) (·400). As in the study of We-
indner et al.,
12
no identification of a lumen was required
for a structure to be considered a lymphatic vessel.
Statistical Analysis
Statistical significance was determined by the v
2
test or
Mann-Whitney’s U-test. Survival rates were calculated
starting from the day of operation. Survival curves
were drawn according to the Kaplan-Meier method, and
differences were analyzed by the log-rank test. The value
P < 0.05 indicates significance. All of the statistical ana-
lyses were performed using StatView 5.0 (Abacus Con-
cepts, Berkeley, CA, USA).
RESULTS
Patient Characteristics
Patients were aged 27 to 85 years (mean – SD,
62 – 13 years), and there were more men (n = 98) than
women (n = 45). After surgery, 123 patients were fol-
lowed from 14 to 2646 days (median 1925 days). Alto-
gether, 33 patients (27%) died of gastric cancer and 7 of
other causes. Thus 83 patients survived, and 40 died.
Immunohistochemistry for VEGF-D Protein
Expression
Typical histologic pictures of VEGF-D staining of par-
affin-embedded specimens are shown in Figure 1. In 55
2 Shida et al.: VEGF-D in Gastric Cancer

patients (39% of the total), VEGF-D protein stained po-
sitive in cancer cell cytoplasm (Fig. 1A) but not in the
surrounding healthy tissue. By contrast, VEGF-D was
undetectable in 88 patients (61%) (Fig. 1B). Staining
lymphatic vessels with D2-40 allowed us to see their
contrast with vascular vessels (Fig. 1C).
VEGF-D Expression Patterns and
Clinicopathologic Variables
Each clinicopathologic variable was compared based
on VEGF-D protein expression (Table 1). Compared with
patients who did not express VEGF-D, those positive for
VEGF-D had the following characteristics in common:
larger tumors, deeper T of TNM, stronger lymphatic
system and venous system invasion, higher LVC, more
metastasis to lymph nodes, M of TNM, and more ad-
vanced stage (Table 1). By contrast, expression of
VEGF-D was not significantly associated with gender,
age, tumor location, histologic grade, or infiltration pat-
tern. The mean (– SD) LVC was 30.0 – 10.9, which was
significantly higher in tumors whose VEGF-D status,
lymphatic invasion, and lymph node metastasis were
positive (Fig. 2).
VEGF-D in Bivariate and Multivariate Analysis
of Survival
Bivariate analysis focusing on relapse-free survival
(RFS) indicated nine significant variables: age, tumor size,
lymphatic system invasion, venous system invasion, pT,
pN, pTNM stage, VEGF-D, and LVC (Table 2). Bivariate
analysis for overall survival (OS) revealed seven signifi-
cant variables: tumor size, lymphatic system invasion,
venous system invasion, pT, pN, pTNM stage, and VEGF-
D (Table 3). The multivariate hazard ratio model identified
VEGF-D expression and pTNM stage as independent
prognostic factors for both RFS and OS (Tables 2, 3).
Even after excluding patients with stage I disease from the
analysis, expression of VEGF-D still exhibited significantly
decreased patient survival for relapse-free survival
(Fig. 3). Although statistical significance was not found, a
tendency toward that significance was observed between
VEGF-D expression and overall survival (Fig. 3).
Figure 1. Immunohistochemical staining for vascular and endothelial grwoth factor-D (VEGF-D) protein in a paraffin-embedded
specimen of gastric cancer. VEGF-D was predominantly expressed in the cytoplasm of cancer cells. Positive (A) and negative (B)
staining of VEGF-D protein in cancer cells. Lymphatic vessels were stained with D2-40 antibody (C). A–C ·400.
Shida et al.: VEGF-D in Gastric Cancer 3

Correlation between VEGF-D Expression and
Recurrence
To determine the relation between VEGF-D expression
and the recurrence pattern, we excluded from further
analysis 20 patients with no recurrence whose follow-up
periods were less than 6 months. Among the remaining
123 patients, 43 had recurrences. The recurrences in-
cluded 7 in lymph nodes, 26 in the peritoneum, and 20 in
distant organs. Furthermore, investigation of the recur-
rent form demonstrated that the frequency of lymph node
recurrence was significantly higher for patients with po-
sitive VEGF-D tumors than in those with negative VEGF-
D tumors (P = 0.0124) (Table 4). A similar correlation
was observed for the frequency of peritoneal recurrence
but not of recurrences in distant organs (Table 4).
DISCUSSION
Tumor metastasis involves a series of complex pro-
cesses in tumor cells that include detachment from the
primary tumor mass, microinvasion into stromal tissues,
intravasation into the lymphatic or blood vessels, and
extravasation and growth at secondary sites.
13,14
There-
Table 1.
Correlation between VEGF-D expression and clinical features
VEGF-D expression
Category Cases Yes No P
Gender
Male 98 37 61 0.9422
Female 45 18 27
Age
< 60 Years 78 34 44 0.2265
‡ 60 Years 65 21 44
Tumor location
U 32 15 17 0.3651
ML 111 40 71
Tumor size
< 4 cm 71 34 37 0.0332
‡ 4cm 72 21 51
Histologic grade
Differentiated 50 18 32 0.7913
Undifferentiated 93 37 56
Growth patterns of tumors infiltrating the surround tissue
IFNa, IFNb 78 26 52 0.2265
IFNc 65 29 36
Lymphatic system invasion
Yes 79 41 38 0.0005
No 64 14 50
Venous system invasion
Yes 63 31 32 0.0299
No 80 24 56
pT
T1 67 16 51 0.0014
T2/T3/T4 76 39 37
pN
Negative 86 22 64 0.0002
Positive 57 33 24
pM
M
0
127 42 85 0.0005
M
1
16 13 3
pTNM stage
I 85 18 67 < 0.0001
II/III/IV 58 37 21
U: upper part; ML: middle or lower part
4 Shida et al.: VEGF-D in Gastric Cancer

fore vascular or lymphatic metastasis may be regulated
by angiogenesis and lymphangiogenesis. Recently, a
significant correlation between lymph node metastasis
and VEGF-C expression has been reported in esopha-
geal, gastric, and colorectal cancers.
15–19
Although
VEGF-D is a potent angiogenic factor in vivo and stimu-
lates endothelial cell proliferation and migration in vitro,
little is known about the physiologic role of this mole-
cule.
20,21
Evidence indicates that VEGF-D stimulates
both angiogenesis and lymphangiogenesis in experi-
mental tumors. On the basis of these data, it has there-
fore been suggested that VEGF-D expression promotes
metastatic vessel growth in tumors.
22
Although a signifi-
cant relation with high-grade VEGF-D expression and a
poor prognosis was reported for colorectal and breast
carcinomas, none has been reported for gastric carci-
noma.
23–25
In the present study, we investigated the
relation between the expression of VEGF-D and the
clinicopathologic status as well as the prognostic signifi-
cance of VEGF-D expression for gastric carcinoma. The
results demonstrated a close association between VEGF-
D expression and tumor size, T of TNM, M of TNM, and
pTNM stage. These results indicate that VEGF-D is ex-
pressed in more advanced tumors. Moreover, lymphatic
and venous system invasion, higher LVCs, and lymph
node metastasis were observed in positive VEGF-D tu-
mors. VEGF-D stimulates lymphangiogenesis and the
dilation of lymphatic vessels by activating the VEGF
receptor-3 (VEGFR-3/Flt-4) expressed in lymphatic
endothelial cells,
26
which in turn causes accelerated
Figure 2. Relation between lymphatic vessel count (LVC) and VEGF-D expression, lymphatic invasion, and lymph node
metastasis. VEGF-D expression is associated with higher LVCs (P < 0.0001); stronger lymphatic system invasion (P = 0.0253); and
more metastasis to lymph nodes (P = 0.0068, Mann-Whitney U-test).
Table 2.
Risk factors affecting relapse-free survival rate determined by Cox proportional hazards model in 126 patients with gastric cancer
Bivariate analysis for relapse-free
survival
Multivariate analysis for relapse-free
survival
Variable Hazard ratio 95% CI P Hazard ratio 95% CI P
Gender (female vs. male) 0.73 0.37–1.45 0.3666
Age (> 60 vs. £ 59 years) 2.20 1.14–4.22 0.0181 1.66 0.82–3.34 0.1566
Tumor location (ML vs. U) 0.93 0.46–1.89 0.8421
Tumor size (> 4.0 cm vs. £ 3.9 cm) 4.72 2.26–9.86 < 0.0001 0.81 0.34–1.97 0.6473
Histologic grade (G
1
vs.G
2
,G
3
) 0.71 0.37–1.39 0.3195
Growth patterns of tumors infiltrating the
surrounding tissue (IFNa/IFNb vs. IFNc)
0.71 0.39–1.29 0.2644
Lymphatic system invasion (yes vs. no) 10.00 3.57–28.05 < 0.0001 0.70 0.21–2.29 0.5582
Venous system invasion (yes vs. no) 4.50 2.26–8.94 < 0.0001 1.37 0.65–2.87 0.4112
pT (T2/T3/T4 vs. T1) 50.41 6.924–367.01 0.0001 5.32 0.63–45.31 0.1261
pN (N
+
vs.N
0
) 9.00 4.15–19.49 < 0.0001 0.70 0.26–1.85 0.4699
pTNM stage (II/III/IV vs. I) 4.65 3.18–6.79 < 0.0001 4.94 2.85–8.65 < 0.0001
VEGF-D (yes vs. no) 5.38 2.75–10.51 < 0.0001 3.31 1.31–8.41 0.0117
LVC (‡ 28 vs. £ 27) 2.32 1.22–4.39 0.0099 0.81 0.36–1.81 0.6033
Shida et al.: VEGF-D in Gastric Cancer 5

lymphatic invasion and lymph node metastasis. Our re-
sults indicate the involvement of VEGF-D-dependent
lymphoangiogenesis and angiogenesis in gastric carci-
noma. Notably, VEGF-D expression and the pTNM stage
were identified as independent indicators for predicting
both RFS and OS. Furthermore, we excluded patients
Table 3.
Risk factors affecting overall survival rate determined by Cox proportional hazards model in 126 patients with gastric cancer
Bivariate analysis for overall survival
Multivariate analysis for overall
survival
Variable Hazard ratio 95% CI P Hazard ratio 95% CI P
Gender (female vs. male) 0.93 0.46–1.87 0.8296
Age (> 60 vs. £59) 1.84 0.94–3.62 0.0771
Tumor location (ML vs. U) 1.20 0.53–2.74 0.6626
Tumor size (‡ 4.0 cm vs. £ 3.9 cm) 3.54 1.71–7.33 0.0007 0.62 0.25–1.51 0.2938
Histologic grade (G
1
vs.G
2
/G
3
) 0.50 0.23–1.10 0.0840
Growth patterns of tumors infiltrating the
surrounding tissue (IFNa/IFNb vs. IFNc)
0.63 0.33–1.21 0.1667
Lymphatic system invasion (yes vs. no) 8.92 3.16–25.24 < 0.0001 0.88 0.26–3.00 0.8382
Venous system invasion (yes vs. no) 4.36 2.10–9.02 < 0.0001 1.56 0.71–3.40 0.2697
pT (T2/T3/T4 vs. T1) 22.32 5.35–93.00 < 0.0001 2.41 0.45–12.96 0.3053
pN (N
+
vs.N
0
) 6.71 3.15–14.28 < 0.0001 0.57 0.22–1.51 0.258
pTNM stage (II/III/IV vs. I) 101.85 13.88–747.27 < 0.0001 4.689 2.71–8.12 < 0.0001
VEGF-D (yes vs. no) 4.12 2.09–8.13 < 0.0001 2.16 1.03–4.54 0.0426
LVC (‡ 28 vs. £ 27) 1.89 0.97–3.67 0.0607
Figure 3. Kaplan-Meier survival analysis of relapse-free survival (RFS) and overall survival (OS) depending on VEGF expression
and pTNM stage. VEGF-D expression still exhibited significantly decreased relapse-free survival (P = 0.0017, log-rank test) after
excluding pTNM stage, although VEGF-D expression did not reach statistical significance for overall survival.
Table 4.
Relation between VEGF-D expression and form of recurrence
Lymph node Peritoneum Distant organ
VEGF-D Yes No Yes No Yes No
Positive 6 41 15 32 7 40
Negative 1 75 11 65 13 63
P 0.0124 0.0378 0.9413
VEGF-D: vascular endothelial growth factor-D.
6 Shida et al.: VEGF-D in Gastric Cancer

with pTNM stage I disease because most of these pa-
tients can be cured by surgery alone. Although expres-
sion of VEGF-D still correlated with significantly
decreased RFS in patients with more than a pTNM stage,
we could only confirm a tendency toward significance for
OS compared to RFS. The simplest explanation for the
result would be that VEGF-D expression did not influence
short-term survival (within a year) (Fig. 3) because our
study included 26 of pTNM stage IV. These results sug-
gest. that VEGF-D expression can be useful for predicting
both OS and RFS, although a larger study is needed to
confirm the prognostic relevance of VEGF-D expression
in gastric cancer. Recently, a number of angiogenesis
inhibitors have undergone clinical testing as single-agent
therapy or in combination with other agents.
27–29
In an
animal model, lymphangiogenesis and lymphatic metas-
tasis were suppressed by a soluble fusion protein,
VEGFR-3-Ig, which inhibits VEGFR-3 signaling.
30
Thus
antiangiogenic therapies targeting VEGF-D may be can-
didates for future cancer therapy.
CONCLUSIONS
Our study demonstrated that expression of VEGF-D is
an independent indicator for RFS and OS in patients with
gastric carcinoma. Moreover, expression of VEGF-D may
be closely correlated with ensuring lymphatic system-
dependent development of this disease. Thus the
assessment of VEGF-D expression could further en-
hance the accuracy of prognosis in patients with gastric
cancer.
REFERENCES
1. Parkin MD, Pisani P, Felay J. Estimate of worldwide inci-
dence of 25 major cancer in 1990. Int J Cancer
1999;80:827–841.
2. Adachi Y, Shiraishi N, Suematsu , et al. Most important
lymh node information on gastric cancer: multivariate
prognostic study. Ann Surg Oncol 2000;7:503–507.
3. Liotta LA. Cancer cell invasion and metastasis. Sci Am
1992;266:54–59, 62–63.
4. Berse B, Brown LF, Van de Water L, et al. Vascular per-
meability factor (vascular endothelial growth factor) gene is
expressed differentially in normal tissues, macrophages,
and tumors. Mol Biol Cell 1992;3:211–220.
5. Ferrara N, Houck K, Jakeman L, et al. Molecular and bio-
logical properties of the vascular endothelial growth factor
family of proteins. Endocr Rev 1992;13:18–32.
6. Veikkola T, Alitalo K. VEGFs, receptors and angiogenesis.
Semin Cancer Biol 1999;9:211–220.
7. Kaipainen A, Korhonen J, Mustonen T, et al. Expression of
the fms-like tyrosine kinase 4 gene becomes restricted to
lymphatic endothelium during development. Proc Natl Acad
Sci U.S.A. 1995;92:3566–3570.
8. Ishikawa M, Kitayama J, Kazama S, et al. Expression of
vascular endothelial growth factor (VEGF-C and –D) is an
important risk factor for lymphatic metastasis in undiffer-
entiated early gastric carcinoma. Jpn J Clin Oncol
2003;33:21–27.
9. Kitadai Y, Kodama M, Cho S, et al. Quantitative analysis of
lymphangiogenic markers for predicting metastasis of hu-
man gastric carcinoma to lymph nodes. Int J Cancer
2005;115:388–392.
10. Japanese Research Society for Gastric Cancer. Japanese
Classification of Gastric Carcinoma, Tokyo, Kanehara,
1999.
11. Kahn HJ, Bailey D, Marks A. Monoclonal antibody D2-40, a
new marker of lymphatic endothelium reacts with Kaposi’s
sarcoma and a subset of angiosarcomas. Mod Pathol
2002;15:434–440.
12. Weidner N, Semple JP, Welch WR, et al. Tumor angio-
genesis and metastasis-correlation in invasive breast car-
cinoma. N Engl J Med 1991;324:1–8.
13. Sleeman JP. The lymph node as a bridgehead in the met-
astatic dissemination of tumors. Recent Results Cancer
Res 2000;157:55–81.
14. Pepper MS. Lymphangiogenesis and tumor metastasis:
myth or reality? Clin Cancer Res 2001;7:462–468.
15. Kitadai Y, Amioka T, Haruma K, et al. Clinicopathological
significance of vascular endothelial growth factor (VEGF)-C
in human esophageal squamous cellcarcinomas. Int J
Cancer 2001;93:662–666.
16. Yonemura Y, Endo Y, Fujita H, et al. Role of vascular
endothelial growth factor C expression in the development
of lymph node metastasis in gastric cancer. Clin Cancer
Res 1999;5:1823–1829.
17. Furodoi A, Tanaka S, Haruma K, et al. Clinical significance
of vascular endothelial growth factor C expression and
angiogenesis at the deepest invasive site of advanced
colorectal carcinoma. Oncology 2002;62:157–166.
18. Akagi K, Ikeda Y, Miyazaki M, et al. Vascular endothelial
growth factor-C (VEGF-C) expression in human colorectal
tissues. Br J Cancer 2000;83:887–891.
19. Amioka T, Kitadai Y, Tanaka S, et al. Vascular endothelial
growth factor-C expression predicts lymph node metastasis
of human gastric carcinomas invading the submucosa. Eur
J Cancer 2002;38:1413–1419.
20. Achen MG, Jeltsh M, Kukk E, et al. Vascular endothelial
growth factor–D (VEGF-D) is a ligand for the tyrosine ki-
nase VEGF receptor 2 (Flk-1) and VEGF receptor 3(Flt-4).
Proc Natl Acad Sci USA 1998;95:548–553.
21. Marconcini L, Marcio S, Morbidelli L, et al. C-fos-induced
growth factor/vascular endothelial growth factor-D induces
angiogenesis in vivo and in vitro. Proc Natl Acad Sci U.S.A.
1999;96:9671–9676.
Shida et al.: VEGF-D in Gastric Cancer 7

22. Stacker SA, Caesar C, Baldwin ME, et al. VEGF-D pro-
motes the metastatic spread of tumor cells via the lym-
phatics. Nat Med 2001;7:186–191.
23. White JD, Hewett PW, Kosuge D, et al. Vascular endothelial
growth factor-D expression is an independent prognostic
marker for survival in colorectal carcinoma. Cancer Res
2002;62:1669–1675.
24. Onogawa S, Kitadai Y, Taknaka S, et al. Expression of
VEGF-C and VEGF-D at the invasive edge correlates with
lymph node metastasis and prognosis of patients with
colorectal carcinoma. Cancer Sci 2004;95:32–39.
25. Nakamura Y, Yasuoka H, Tsujimoto M, et al. Prognostic
significance of vascular endothelial growth factor D in
breast carcinoma with long-term follow-up. Clin Cancer Res
2003;9:717–721.
26. Joukov V, Sorsa T, Kumar V, et al. Proteolytic processing
regulates receptor specificity and activity of VEGF-C.
EMBO J 1997;16:3898–3911.
27. Yang JC, Haworth L, Sherry RM, et al. A randomized trial of
bevacizumab, an anti-vascular endothelial growth factor
antibody, for metastatic renal cancer. N Engl J Med
2003;349:427–434.
28. Willett CG, Boucher Y, di Tomaso E, et al. Direct evidence
that the VEGF-specific antibody bevacizumab has anti-
vascular effects in human rectal cancer. Nat Med
2004;10:145–147.
29. Zangari M, Anaissie E, Stopeck A, et al. Phase II study of
SU5416, a small molecule vascular endothelial growth
factor tyrosine kinase receptor inhibitor, in patients with
refractory multiple myeloma. Clin Cancer Res 2004;10:88–
95.
30. He Y, Kozaki K, Karpanen T, et al. Suppression of tumor
lymphangiogenesis and lymph node metastasis by blocking
vascular endothelial growth factor receptor 3 signalling. J
Natl Cancer Inst 2002;94:819–825.
8 Shida et al.: VEGF-D in Gastric Cancer