Quantitative Analysis of Lymphangiogenic Markers in Human Gastroenteric Tumor

Archives of Medical Research 38 (2007) 106e112

ORIGINAL ARTICLE

Quantitative Analysis of Lymphangiogenic Markers in Human Gastroenteric Tumor Lu Yuanming,* Gao Feng,* Tao Lei, and Wang Ying Clinical Research Center, No. 6 Hospital of Shanghai Jiaotong University, Shanghai, People’s Republic of China Received for publication May 19, 2006; accepted August 4, 2006 (ARCMED-D-06-00206).

Background. Lymphatic spread of gastroenteric tumor cells to regional lymph nodes is one of the early events in metastatic cancers and is often associated with distant metastatic spread and poor prognosis. Expression levels of newly described lymphatic endothelial markers, LYVE-1, VEGF-C, VEGF-D and the VEGF receptors VEGFR-3 were assessed in our study. Methods. Paired (tumor and corresponding normal tissue) samples were obtained. The expression level of each factor was determined using RT-PCR and quantified by using a real-time quantitative PCR (RT-QPCR) technique, with respective cloned cDNA plasmids as internal standards. Results. The expression of VEGF-C and lymphatic endothelial marker VEGFR-3 was significantly greater in patients with lymph node metastasis than in those without metastasis, but no different expression level of VEGF-D and LYVE-1 was detected in both groups of patients. Lymphatic vessel density (LVD), which was assessed by immunohistochemistry for LYVE-1, was correlated with lymphangiogenesis factors and lymph node metastasis. Expression of VEGF-C and VEGFR-3 was significantly associated with higher peritumoral LVD than normal group, and LVD was found greater in the node-positive group than in the node-negative group. Conclusions. These results indicate that quantitative analysis of lymphangiogenic marker VEGF-C and VEGFR-3 in gastroenteric specimens may be useful in predicting metastasis of gastroenteric cancer to regional lymph nodes, but the role of LYVE-1 in predicting metastasis of gastroenteric cancer requires further analysis. Ó 2007 IMSS. Published by Elsevier Inc. Key Words: Gastroenteric tumor, Lymphangiogenesis, Vascular endothelial growth factor (VEGF), Lymphatic vessel endothelial HA receptor-1 (LYVE-1).

Introduction Lymphangiogenesis refers to the formation of new lymphatic vessels that may occur in normal developing tissues or in tumors. Once new lymphatic communications have been established, many malignant tumors spread via the lymphatics as well as other routes (1). Lymphatic spread of gastric cancer cells to regional lymph nodes is one of

*These authors contributed equally to this work. Address reprint requests to: Lu Yuanming, Clinical Research Center, No. 6 Hospital of Shanghai Jiaotong University, 600 Yi-shan Road, Shanghai, 200233, People’s Republic of China; E-mail: luym650920@hotmail. com

the early events in metastatic cancers and is often associated with distant metastatic spread and poor prognosis. The extent of regional lymph node metastasis is an important indicator of tumor aggressiveness and is a prognostic factor for patients with gastric tumor (2). Recently, many factors have been proposed as specific markers for developing lymphatic endothelium. Vascular endothelial growth factor (VEGF), a member of the platelet-derived growth factor family, is a major inducer of angiogenesis and vessel permeability (3). VEGF-C and VEGF-D are ligands for VEGFR-3 (also termed fms-like tyrosine kinase 4, Flt-4), a tyrosine kinase receptor that is expressed predominantly in lymphatic endothelial cells (4). Overexpression of VEGF-C or VEGF-D can lead to

0188-4409/06 $esee front matter. Copyright Ó 2007 IMSS. Published by Elsevier Inc. doi: 10.1016/j.arcmed.2006.08.009

Lymphangiogenesis in Gastroenteric Tumor

lymphangiogenesis, intralymphatic tumor growth and formation of lymph node metastases (5e7). Both VEGF-C and VEGF-D predominantly act on VEGFR (8e9). The VEGFR-3 receptor is mainly expressed on the lymphatic endothelium of adult tissues (10e11) and is increased in a variety of tumor types (12e13). LYVE-1 is a factor considered to be specific for lymphatic endothelium. LYVE-1 is a lymphatic endothelial receptor for the extracellular matrix/lymphatic fluid glycosaminoglycan, hyaluronan (14). LYVE-1-expressing vessels were positive also for the VEGFR-3 lymphatic marker (15). However, results regarding the correlations between the expression of VEGF-C, D, VEGFR-3, and LYVE-1 in human gastroenteric tumor are still uncertain. For example, Stefan et al. (16) reported VEGF-D and VEGFR-3, but not VEGF-C, as independent prognostic parameters in gastric adenocarcinoma. But in the report of Yasuhiko et al. (17), the conclusion is that VEGF-C, but not VEGF-D, appears to be involved in lymph node metastasis of gastric carcinoma. Quantification of VEGF-C, VEGFR-3 mRNAs in biopsy specimens by real-time PCR may prove to be useful in predicting lymph node metastasis. In our study, we first quantified mRNA expression levels of VEGF-C, VEGF-D, VEGFR-3, and LYVE-1 in biopsy specimens of human gastroenteric tumor and correlated these levels with lymph node metastasis and tumor progression. We then studied whether the VEGF family factors and LYVE-1 can serve as prognostic marker for lymph node metastasis. Materials and Methods

107

was performed with the isolated RNA (1 mg). The primers for VEGF-C, VEGF-D, VEGFR-3, LYVE-1 and B-actin are given in Table 1. The specificity of the primer sequences was confirmed by FASTA (EMBL database). RT-PCR was performed with extracted RNA and oligomers as templates and primers, respectively. cDNA was amplified by 30 cycles of denaturation for 30 sec at 94 h, annealing for 30 sec at 55e62 h based on the primer, and extension for 3 min at 72 h. At the end of the reaction, the mixtures were loaded onto a 2% agarose gel and stained with ethidium bromide prior to examination under UV light. Cloning of Lymphangiogenic Factors into Plasmids Lymphangiogenic factors were amplified from the colorectal cDNA using the primer pairs in Table 1. Appropriate PCR products were then cloned using the pCR2.1-TOPO vector and amplified using One-Shot E. coli (Invitrogen, Carlsbad, CA). Bacterial colonies were grown on X-galcoated LB agar plates, and white colonies were examined via PCR, using the primer pairs from Table 1 to confirm the presence of the appropriate factor. Plasmid purification was performed by using the plasmid mini-purification kit (Qiagen). Plasmids were then sequenced to confirm that they contained the correct DNA sequences. The concentration of each plasmid was calculated using a spectrophotometer, copy number was calculated based on the size of plasmid and concentration, and serial logarithmic dilutions were prepared. These calculated plasmid dilutions were used as internal standards.

Samples

Quantitative Real-time RT-PCR Analysis

Paired (tumor and corresponding normal mucosa) tissue samples were obtained from gastroenteric cancer patients (n 5 37) and snap frozen in liquid nitrogen. During surgical removal of each tumor, an adjacent section of normal mucosa was also removed for normal background tissue, following pathological confirmation that it was free from tumor deposits. Informed consent was obtained from all patients for participation in the study. Pathology reports and clinical histories were reviewed for accurate staging at the time of surgery. Lymph node status was determined by routine pathological examination with the surgical specimens. Two groups of patients, those with lymph node metastasis (node-positive group, n 5 20) and those without (node-negative group, n 5 17) were closely matched for histological type and depth of invasion. All patients had invasive gastroenteric carcinoma in which the tumor invasion was beyond the submucosa.

PCR reactions were performed in an ABI Prism 7000 (Applied Biosystems, Foster City, CA) and expression levels (copies/mL from internal standard) of lymphangiogenic factors were quantified. After 10 min at 95 h to denature the cDNA and to activate the Taq DNA polymerase, the cycling conditions were as follows: 40 cycles of denaturation at 95 h for 15 sec, annealing at 55e62 h based on the different primer (Table 2) for 5 sec, and extension at 72 h for 12 sec. To ensure that the correct products were amplified, Table 1. Primer sequences for semiquantitative RT-PCR Primer sequences LYVE-1 VEGF-C VEGF-D

Semiquantitative RT-PCR Total RNA was extracted from gastroenteric carcinoma specimens with an RNeasy Mini Kit (Qiagen, Valencia, CA) according to the manufacturer’s instructions. RT-PCR

VEGFR-3 (FLT4) b-actin

F: 50 -ATCCCATATTCAACACTCAA-30 R: 50 -CCTTTACTACTTTGGTTTCG-30 F: 50 -CACTGCCACAAGAAGTACC-30 R: 50 -CCAGAAGAAGACAGCGAT-30 F: 50 -GAGGAGCAGTTACGGTCTGT-30 R: 50 -GTAGCTCGTGCTGGTGTTCA-30 F: 50 -GTATGGACTCTCGCTCAGCAT-30 R: 50 -AGGCTCTCTTCATTGCAACAG-30 F: 50 -TGCATTGTTACAGGAAGTCCCTT-30 R: 50 -GGGAGAGGACTGGGCCAT-30

Yuanming et al./ Archives of Medical Research 38 (2007) 106e112

108

Table 2. Primer and probe sequences for quantitative RT-PCR 0

Sequence (5 Fam-3 Tamra)

Direction LYVE-1

Forward Reverse Probe VEGF-C Forward Reverse Probe VEGF-D Forward Reverse Probe VEGFR-3 Forward (FLT4) Reverse Probe b-actin Forward Reverse Probe

0

50 -GCCTGGTGTTGCTTCTCACTT-30 50 -GTGATCCCCATAATTCTGCATGA-30 50 -TGCGTGCAGAAGAGCTTTCCATCCA-30 50 -GTTCCACCACCAAACATGCA-30 50 -CACTATATGAAAATCCTGGCTCACAA-30 50 -ACGGCCATGTACGAACCGCCA-30 50 -CTGGAACAGAAGACCACTCTCATC-30 50 -CTCGCAACGATCTTCGTCAAA-30 50 -CAGGAACCAGCTCTCTGTGGGC-30 50 -TCGCGCGGTTCTCGAA-30 50 -TTCGGGAAGCCAGGAACTC-30 50 -TCTCCAGACCAAGAAGCTGAGGACCTGT-30 50 -CCGTCTTCCCCTCCATCG-30 50 -GTCCCAGTTGGTGACGATGC-30 50 -CCAGGGCGTGATGGTGGGCAT-30

all samples were separated by 2% agarose gel electrophoresis. The numbers of copies in unknown samples were calculated by comparison of their Cps with the internal standard curve, and the level of cDNA (copies/mL) in the cancer samples was calculated. Internal control of b-actin was also performed on the same samples to correct for any residual differences in the initial level of RNA in the specimens by expressing the amount of cDNA as a ratio to the amount of b-actin. Immunohistochemistry Consecutive 4-mm-thick sections were cut from each paraffin sample. Sections were immunolabeled for LYVE-1 and CD34. Immunohistochemical labeling was performed by the immunoperoxidase method following the antigen retrieval with 0.1% trypsin (37 h, 30 min). Antibodies used were a mouse monoclonal antibody for LYVE-1 (AngioBio, Del Mar, CA) and a rabbit polyclonal antibody for CD34 (Dako Cytomation, Glostrup, Denmark). Negative controls were performed with nonspecific IgG as the primary antibody. Immunohistochemistry was carried out with an LSAB Kit (Dako). Lymphatic vessel density and microvessel density were determined from the counts of LYVE1-positive vessels and CD34-positive/LYVE-1-negative vessels, respectively. Vessel density was assessed by light microscopy of the intratumoral region containing the greatest number of capillaries and small venules. Highly vascular areas were identified by scanning tumor sections at low power (40 and 100). After the six areas of greatest neovascularization were identified, a vessel count was performed at 200, and the mean count of six fields was calculated. In slides immunolabeled for LVYE-1, only vessels with typical morphology (including a lumen) were counted as lymphatic vessels because of occasional weak antibody cross-reactivity with fibroblasts.

Statistical Analysis Statistical significance was determined by the unpaired Student’s t-test and Spearman’s correlation coefficient. For levels of samples, b-actin was expressed as a ratio of the respective initial concentrations. The difference between node positive and negative samples was also analyzed.

Results Expression of VEGF-C, VEGF-D, VEGFR-3, and LYVE-1 mRNA in Gastroenteric Cancer Samples We initially examined the expression of VEGF-C, VEGFD, VEGFR-3, and LYVE-1 mRNAs in gastroenteric carcinoma by semiquantitative RT-PCR. Different expression level of mRNAs for VEGF-C, VEGF-D, VEGFR-3, and LYVE-1 was detected in gastroenteric tumor samples (Figure 1). b-actin was used as internal control in semiquantitative RT-PCR. Relation between Expression of VEGF-C, VEGF-D, VEGFR-3, and LYVE-1 mRNA, and Nodal Metastasis in Human Gastroenteric Cancer We next examined the mRNA expression of the potent lymphangiogenic factors by quantitative real-time PCR. The relative expression levels of lymphangiogenic factors are shown according to node status in Figure 2. Patients with positive lymph nodes showed significantly greater expression of VEGF-C but not of VEGF-D than shown by node-negative patients. We also examined expression of the lymphatic markers VEGFR-3 and LYVE-1 and found that the expression levels of VEGFR-3 were also significantly greater in the node-positive group than in the node-negative group, but the expression levels of LYVE-1 are too low to compare with other factors. Immunohistochemistry for LYVE-1 and CD34 in Human Gastroenteric Cancer Samples Immunohistochemistry for LYVE-1 and CD34 was carried out to examine the relation between LVD (lymphatic vessel density) and MVD (microvessel density) and regional lymph node metastasis. LYVE-1 positive vessels were

Figure 1. Semiquantitative RT-PCR for expression of VEGF-C, VEGFR-3, VEGF-D and LYVE-1 in gastroenteric tumor samples. b-actin was applied to internal control. (N, Normal; C, Cancer; 1, VEGFR-3; 2, VEGF-D; 3, LYVE-1; 4, VEGF-C).

Lymphangiogenesis in Gastroenteric Tumor

A

150

**

109

B

100

mRNA level (copies/µL)

mRNA level (copies/µL)

30

50

10

0

0 normal

C

normal

node-positive node-negative

**

D

node-positive node-negative

2

200

1.5

mRNA level (copies/µL)

150

mRNA level (copies/µL)

20

100

1

0.5

50

0 normal

node-positive node-negative

0 normal

node-positive node-negative

Figure 2. Correlation of VEGF-C (A), VEGF-D (B), VEGFR-3 (C), and LYVE-1 (D) mRNA expression with regional lymph node metastasis in human gastroenteric tumor samples. There was a significant increase in VEGF-C and VEGFR-3 expression in the cancer samples (**p !0.01) compared with the normal samples. The different expression levels of node-positive and node-negative were also detected in both of these two factors (**p !0.01). , node-positive cancer sample, node-negative cancer sample, - normal tissue.

identified in gastroenteric tumor tissues but were rare in normal mucosa (Figure 3). CD34 immunoreactivity was identified predominantly in blood microvessels. LVD was found correlated with lymphangiogenesis factors and lymph node metastasis. The expression of VEGF-C and VEGFR-3 (but not VEGF-D and LYVE-1) was significantly associated with higher peritumoral LVD (Figure 4). LVD was found greater in the node-positive group than in the node-negative group (16.9  1.6 vs. 4.2  1.3; p !0.01). However, CD34-positive/LYVE-1 negative MVD was not associated with lymph node metastasis (node-positive vs. node-negative; 45.7  2.3 vs. 47.2  3.1) (Figure 5).

Discussion The mechanisms of the lymphatic spread of tumor cells remain unclear at present. It is unclear if tumors can induce lymphangiogenesis or whether cells just invade through existing peritumor vessels. It is suggested that the lymphatic system as a functional entity is not corroborated within tumors because of the increased interstitial pressure created by the proliferating cancer cells (18). Recent studies indicated that the centers of tumors do not contain functional lymphatics; however, lymphatic vessels at the tumor margin do facilitate lymphatic spread of tumor cells (19). This

110

Yuanming et al./ Archives of Medical Research 38 (2007) 106e112

Figure 3. Immunohistochemical labeling for LYVE-1 (A,B) and CD34 (C) in the normal gastric mucosa and in primary gastric carcinomas. LYVE-1-positive vessels were identified in gastric carcinoma tissues (B) but were rare in normal mucosa (A). CD34 immunoreactivity was identified predominantly in blood microvessels (C).

pattern is that, with related factors that modulate lymphangiogenesis, it is new lymphatic vessel formation at the periphery of the tumor that enhances metastatic spread rather than growth from the center of the tumor (20). Overexpression of VEGF-C and VEGF-D results in the enlargement of peritumor lymphatic vessels and increased intratumor lymphatic density and is significantly associated with an increase of metastases to the regional lymph nodes (17,21). In our study, overexpression of VEGF-C but not VEGF-D was detected in gastroenteric tumor samples. Correlation of VEGF-C, VEGF-D and/or VEGFR-3 with lymphatic spread, tissue invasion and/or poor prognosis has been observed in colorectal, endometrial, ovarian and breast cancer (22e25). High density of VEGFR-3-positive vessels was found to correlate with poor prognosis in endometrial, nonsmall-cell lung carcinoma and breast cancer (17,26,27). In other studies, however, such correlations could not be confirmed (28,29) or opposite relationships

were found (30,31), indicating that effects and interaction of the VEGF-C/VEGF-D/VEGFR-3 system in cancer biology are complex and may differ among malignancies (32). Our results reveal that the VEGFR-3, the receptor to VEGF-C and VEGF-D, was expressed at a significantly higher level in gastroenteric tumor samples than in normal background tissues. This suggests that VEGF-C may exert a more powerful effect to enhance lymphangiogenesis and facilitate lymphatic dissemination of tumor cells in gastroenteric tumor. Invasion to lymphatic vessels and metastasis to lymph nodes are frequent complications in human gastroenteric tumor. To clarify the mechanism of its occurrence, correlation of VEGF-C, VEGF-D, VEGFR-3, and LYVE-1 mRNA expression and lymphatic vessel density (LVD) was also studied in our investigation. Expression of VEGF-C, VEGFR-3 was significantly associated with the higher peritumoral LVD and lymph node metastasis in tumor,

Lymphangiogenesis in Gastroenteric Tumor

LVD (mean±SE)

LVD (mean±SE)

20

**

10

10

tumor tissue

0

normal tissue

tumor tissue

normal tissue

10

0

** 20

15

5

5

5

D

15

15

15

0

C

20

LVD (mean±SE)

B

**

20

LVD (mean±SE)

A

111

10

5

0 tumor tissue

normal tissue

tumor tissue

normal tissue

Figure 4. Correlation of mRNA level of lymphangiogenesis factors VEGF-C (A), VEGF-D (B), VEGFR-3 (C), LYVE-1 (D) with lymphatic vessel density (LVD) in the primary tumor. **p !0.01.

and these finding suggest that VEGF-C promotes the proliferation of peritumoral lymphatic vessels and that lymphatic invasion and metastasis to lymph nodes are frequently induced in human gastroenteric tumor. LYVE-1 is a novel and specific lymphatic endothelial marker expressing on the lymph vessel wall as epithelial receptor molecule for hyaluronic acid. LYVE-1 expresses highly in lymph vessels and is completely absent from blood vessels (17). The level of LYVE-1 was studied in our experiment but the result was uncertain. No difference was found in the mRNA expression of LYVE-1 in tumor and normal tissue. LYVE-1 stained positive in non-functional intratumor vessels as well as the lymphatic vessels peritumor although mainly on the periphery of the tumor. Based on these results, the value of LYVE-1 was questioned in identifying fully functional lymphatics.

A

B

80

30 60

MVD (mean±SE)

LVD (mean±SE)

25 20 15 10

40

20

5 0

0 node-positive

node-negative

node-positive

node-negative

Figure 5. Correlation of LVD and MVD with regional lymph node metastasis in human gastroenteric cancer samples (mean  SE). (A) Lymphatic vessel density (LVD) was significantly greater in the node-positive group and negative group (**p !0.01). (B) No difference of microvessel density (MVD) was found in both of node-positive and node-negative group ( p O0.05).

References 1. Stacker SA, Achen MG, Jussila L, Baldwin ME, Alitalo K. Lymphangiogenesis and cancer metastasis. Nat Rev Cancer 2002;2:573e583. 2. Adachi Y, Shiraishi N, Suematsu T, Shiromizu A, Yamaguchi K, Kitano S. Most important lymph node information in gastric cancer: multivariate prognostic study. Ann Surg Oncol 2000;7:503e507. 3. Ferrara N, Houck K, Jakeman L, Leung DW. Molecular and biological properties of the vascular endothelial growth factor family of proteins. Endocr Rev 1992;13:18e32. 4. Kaipainen A, Korhonen J, Mustonen T, van Hinsbergh VW, Fang GH, Dumont D, et al. Expression of the fms-like tyrosine kinase 4 gene becomes restricted to lymphatic endothelium during development. Proc Natl Acad Sci USA 1995;92:3566e3570. 5. Jeltsch M, Kaipainen A, Joukov V, Meng X, Lakso M, Rauvala H, et al. Hyperplasia of lymphatic vessels in VEGF-C transgenic mice. Science 1997;276:1423e1425. 6. Yonemura Y, Endo Y, Fujita H, Fushida S, Ninomia I, Bandou E, et al. Role of vascular endothelial growth factor C expression in the development of lymph node metastasis in gastric cancer. Clin Cancer Res 1999;5:1823e1829. 7. Stacker SA, Caesar C, Baldwin ME, Thornton GE, Williams RA, Prevo R, et al. VEGF-D promotes the metastatic spread of tumor cells via the lymphatics. Nat Med 2001;7:186e191. 8. Achen MG, Jeltsch M, Kukk E, Makinen T, Vitali A, Wilks A, et al. Vascular endothelial growth factor-D (VEGF-D) is a ligand for the tyrosine kinases VEGF receptor 2 and VEGF receptor 3. Proc Natl Acad Sci USA 1998;95:548e553. 9. Joukov V, Sorsa T, Kumar V, Jeltsch M, Claesson-Welsh L, Cao Y, et al. Proteolytic processing regulates receptor specificity and activity of VEGF-C. EMBO J 1997;16:3898e3911. 10. Kaipainen A, Korhonen J, Mustonen T, van Hinsbergh VWM, Fang G-H, Dumont D, et al. Expression of the fms-like tyrosine kinase 4 gene becomes restricted to lymphatic endothelium during development. Proc Natl Acad Sci USA 1995;92:3566e3570. 11. Partanen T, Arola J, Saaristo A, Jussila L, Ora A, Miettinen M, et al. VEGF-C and VEGF-D expression in neuroendocrine cells and their receptor, VEGFR-3, in fenestrated blood vessels in human tissues. FASEB J 2000;14:2087e2096. 12. Yonemura Y, Endo Y, Tabata K, Kawamura T, Yun HY, Bandou E, et al. Role of VEGF-C and VEGF-D in lymphangiogenesis in gastric cancer. Int J Clin Oncol 2005;10:318e327. 13. Schneider M, Buchler P, Giese N, Giese T, Wilting J, Buchler MW, et al. Role of lymphangiogenesis and lymphangiogenic factors during pancreatic cancer progression and lymphatic spread. Int J Oncol 2006; 28:883e890.

112

Yuanming et al./ Archives of Medical Research 38 (2007) 106e112

14. Banerji S, Ni J, Wang SX, Su J, Tammi R, Jones M, et al. LYVE-1, a new homologue of the CD44 glycoprotein, is a lymph-specific receptor for hyaluronan. J Cell Biol 1999;144:789e801. 15. Skobe M, Hawighorst T, Jackson DG, Prevo R, Janes L, Velasco P, et al. Induction of tumor lymphangiogenesis by VEGF-C promotes breast cancer metastasis. Nat Med 2001;7:192e198. 16. Stefan J, Christoph W, Thomas J, Michael V, Johannes H, Stephan G, et al. Vascular endothelial growth factor-D and its receptor VEGFR-3: Two novel independent prognostic markers in gastric adenocarcinoma. J Clin Oncol 2006;24:228e240. 17. Yasuhiko K, Michiyo K, Songde C, Tsuyoshi K, Takehiko O, Shigeru K, et al. Quantitative analysis of lymphangiogenic markers for predicting metastasis of human gastric carcinoma to lymph nodes. Int J Cancer 2005;115:388e392. 18. Pepper MS. Lymphangiogenesis and tumour metastasis: myth or reality? Clin Cancer Res 2001;7:462e468. 19. Padera TP, Kadambi A, Thomaso E Di, Carreira CM, Brown EB, Boucher Y, et al. Lymphatic metastasis in the absence of functional intra tumour lymphatics. Science 2002;296:1883e1886. 20. Gershenwald JE, Fidler IJ. Targeting lymphatic metastasis. Science 2002;296:1811e1812. 21. Mandriota SJ, Jussila L, Jeltsch M, Compagni A, Baetens D, Prevo R, et al. Vascular endothelial growth factor-C-mediated lymphangiogenesis promotes tumour metastasis. EMBO J 2001;20:672e682. 22. White JD, Hewett PW, Kosuge D. Vascular endothelial growth factorD expression is an independent prognostic marker for survival in colorectal carcinoma. Cancer Res 2002;62:1669e1675. 23. Onogawa S, Kitadai Y, Tanaka S. 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;95:32e39. 24. Yokoyama Y, Charnock-Jones DS, Licence D. Expression of vascular endothelial growth factor (VEGF)-D and its receptor, VEGF receptor

25.

26.

27.

28.

29.

30.

31.

32.

3, as a prognostic factor in endometrial carcinoma. Clin Cancer Res 2003;9:1361e1369. Nakamura Y, Yasuoka H, Tsujimoto M. Prognostic significance of vascular endothelial growth factor d in breast carcinoma with long-term follow-up. Clin Cancer Res 2003;9:716e721. Arinaga M, Noguchi T, Takeno S. Clinical significance of vascular endothelial growth factor C and vascular endothelial growth factor receptor 3 in patients with nonsmall cell lung carcinoma. Cancer 2003;97:457e464. Nakamura Y, Yasuoka H, Tsujimoto M. Flt-4-positive vessel density correlates with vascular endothelial growth factor-d expression, nodal status, and prognosis in breast cancer. Clin Cancer Res 2003;9: 5313e5317. Gunningham SP, Currie MJ, Han C. The short form of the alternatively spliced flt-4 but not its ligand vascular endothelial growth factor C is related to lymph node metastasis in human breast cancers. Clin Cancer Res 2000;6:4278e4286. Jacquemier J, Mathoulin-Portier MP, Valtola R. Prognosis of breastcarcinoma lymphagenesis evaluated by immunohistochemical investigation of vascular-endothelial-growth-factor receptor 3. Int J Cancer 2000;89:69e73. Charoenrat PO, Rhys-Evans P, Eccles SA. Expression of vascular endothelial growth factor family members in head and neck squamous cell carcinoma correlates with lymph node metastasis. Cancer 2001; 92:556e568. Niki T, Iba S, Tokunou M. Expression of vascular endothelial growth factors A, B, C, and D and their relationships to lymph node status in lung adenocarcinoma. Clin Cancer Res 2000;6:2431e2439. Hung CJ, Ginzinger DG, Zarnegar R, Kanauchi H, Wong MG, Kebebew E, et al. Expression of vascular endothelial growth factor C in benign and malignant thyroid tumors. J Clin Endocrinol Metab 2003;88:3694e3699.