Article Tools
Gastrointestinal Cancer
Article Tools
Vascular Endothelial Growth Factor-D and Its Receptor VEGFR-3: Two Novel Independent Prognostic Markers in Gastric Adenocarcinoma
Abstract
Vascular endothelial growth factor (VEGF)-D and its homolog VEGF-C influence lymphangiogenesis through activation of VEGF receptor 3 (VEGFR-3), and have been implicated in lymphatic tumor spread. Nodal dissemination of gastric adenocarcinomas critically determines clinical outcome and therapeutic options of affected patients. Therefore, we analyzed expression and prognostic significance of VEGF-D along with VEGF-C, and VEGFR-3 in gastric adenocarcinomas.
VEGF-C, VEGF-D, and VEGFR-3 were analyzed in 91 R0-resected primary gastric adenocarcinomas, corresponding noncancerous gastric mucosa, and lymph node metastases employing immunohistochemistry and/or in situ hybridization. Blood and lymph vessel densities were assessed after staining with CD31 and LYVE-1–specific antibodies.
VEGF-D and VEGF-C were detected in 67.0% and 50.5% of gastric cancers, respectively. Healthy gastric mucosa was negative for VEGF-C and in 12.5% positive for VEGF-D. Presence of VEGF-D (P = .005) or VEGF-C (P = .006) was correlated with lymphatic metastases and decreased survival (VEGF-D, P < .05; VEGF-C, P < .05). VEGFR-3 was correlated with reduced carcinoma-specific survival (P < .05), and Cox multivariate regression analysis qualified VEGF-D and VEGFR-3, but not VEGF-C, as independent prognostic parameters. In lymph node–positive gastric cancers, presence of VEGF-D/VEGFR-3 was associated with poor survival, whereas absence of VEGF-D/VEGFR-3 defined a subgroup of patients with clearly favorable prognosis.
VEGF-D and VEGFR-3 are novel independent prognostic marker molecules aiding to identify patients with poor prognosis after curative resection of gastric adenocarcinomas. Combined analysis of the VEGF-C/VEGF-D/VEGFR-3 system can be useful to identify patients with unfavorable clinical outcome and thereby may help to refine therapeutic decisions in gastric cancer.
Gastric adenocarcinoma is the second leading cause of cancer-related death worldwide, displaying an annual incidence of 80 to 90 deaths per 100,000 cases in Japan and 5 to 15 deaths 100,000 cases in Europe.1 In the Far East, most gastric cancer patients are diagnosed with early gastric cancers,1 which are limited to the gastric mucosa and submucosa, rarely show metastatic spread, and display 10-year survival rates between 80% and 95%.1 In Western countries, patients are diagnosed mostly at advanced clinical stages, typically showing lymphatic tumor dissemination and a poor prognosis with 5-year survival rates less than 30%.2 Clinical studies demonstrated that the extent of lymphatic dissemination critically determines the clinical outcome of gastric cancer patients, and that surgical clearance of lymphatics is essential if curative treatment is intended.2,3 Therefore, exact determination of nodal spread is indispensable for optimizing therapeutic strategies and correct assessment of patients prognosis. Current procedures for detection of lymph node affection in gastric cancer include computed tomography (CT) scan, endoluminal ultrasound, and laparascopy.2,4 Although clinically well established, these methods also have clear limitations.2 Therefore, identification of molecular markers for assessment of nodal status and prognosis in gastric cancer patients is highly desirable.
Angiogenesis, the formation of new blood vessels from endothelial precursors, is a prerequisite for growth and progression of solid malignancies,5 and the vascular endothelial growth factor (VEGF) superfamily of endothelial growth factors has been identified to critically influence tumor-related angiogenesis.5,6 Recently, it became clear that basic mechanisms of hemangiogenesis may also apply to the lymphatic system, and that VEGF-C and its close homolog VEGF-D are intimately involved in the regulation of lymphangiogenic processes.7 VEGF-C and VEGF-D exert their effects on endothelial cells via activation of VEGF receptor (VEGFR)-2 and VEGFR-3 (also termed fms-like tyrosine kinase 4 [flt4]).7-10 Importantly, full-length VEGF-C and VEGF-D isoforms display high affinity for VEGFR-3, whereas their affinities for VEGFR-2 is increased through progressive proteolytic cleavage by tissue proteases including plasmin.11-13 In adult endothelium, VEGFR-3 expression is predominantly restricted to lymphatics,9 but has also been detected on blood vessels of malignant tumors and during wound healing.10,14 In contrast, VEGFR-2 is mainly expressed by blood vessel endothelial cells, where it signals for vasculogenesis and angiogenesis.9,12
In clinical studies, correlation of VEGF-C, VEGF-D and/or VEGFR-3 with lymphatic spread, tissue invasion and/or poor prognosis has been observed in colorectal,15,16 endometrial,17 ovarian18 and breast cancer.19 Similarly, high density of VEGFR-3–positive vessels was found to correlate with poor prognosis in endometrial,17 non–small-cell lung carcinoma20 and breast cancer.21 In other studies, however, such correlations could not be confirmed,22,23 or opposite relationships were found,24,25 indicating that effects and interaction of the VEGF-C/VEGF-D/VEGFR-3 system in cancer biology are complex and may differ between malignancies. In gastric cancer, a correlation with poor prognosis has been found for VEGF-C,26,27 but no reports on the importance of VEGFR-3 in this context are available.28-31 Moreover, no study investigating the expression of VEGF-D and its relationship to the clinical outcome of gastric cancer has yet been published. Therefore, we analyzed the abundance of VEGF-D along with VEGF-C and VEGFR-3 in gastric adenocarcinomas, and correlated the findings with clinicopathologic parameters and patient survival.
All tissues investigated in this study were obtained from patients (n = 91) who underwent curative (R0) gastrectomy between 1995 and 2003 at the Division of Surgery and Surgical Oncology, Robert Rössle Hospital, Charité, Campus Buch (Berlin, Germany). Written informed consent for gene expression analyses and experimental immunohistochemistry was obtained from all patients before analyses. Whereas abundance of VEGF-C and VEGF-D was analyzed in all patients included, assessment of VEGFR-3 was conducted in 88 of 91 cases because of limited availability of tumor material. Patients’ ages ranged from 34 to 85 years (mean, 62.7 years; SE, 11.2 years). Four patients were lost to follow-up and thus censored at the time of last contact (mean follow-up time, 37.7 months; SE, 27.3 months). In one patient, the cause of death was unknown, and the death of another six patients was not related to gastric cancer. Thirty-nine patients showed disease relapse and 34 patients had died at the end of the study as a result of gastric cancer. Among the patients analyzed, two individuals received neoadjuvant chemotherapy with fluorouracil, cisplatin, and folinic acid. Exclusion of these patients from survival analysis did not influence results. No other neoadjuvant and/or adjuvant therapy (chemo- and/or radiotherapy) was applied to patients included on this study. Staging and diagnosis of gastric carcinoma was assessed according to the WHO classification32 and the TNM classification set out by the International Union Against Cancer (Union International Contre le Cancer [UICC]).33
Gastric cancer cell lines were cultured as previously described.34 Endothelial cell line HMEC-1 was kept in MCDB-131 medium supplemented with 15% bovine calf serum (Biochrom KG, Berlin, Germany), 10 ng/mL epidermal growth factor (R&D, Wiesbaden Norderstedt, Germany) and 1 μg/mL hydrocortisone. Human umbilical vein endothelial cells (HUVECs) were cultured on 0.2% gelatin-coated dishes (Sigma, Schnelldorf, Germany) and maintained in 199 Earle Medium (Biochrome KG) supplemented with 20% bovine calf serum (Biochrom KG) and 1% retina-derived growth factor. Total RNA was isolated using the TRIZOL reagent (Invitrogen, Karlsruhe, Germany) and reverse transcribed with Oligo-dT primers and SuperScript II (Invitrogen). cDNAs generated from 50 ng of total RNA were used for real-time quantification using gene-specific primers (Table 1) and the Brilliant QPCR Kit (Stratagene, Amsterdam, the Netherlands) on a Stratagene MX3000P cycler. Quantification was performed by the comparitve ΔCT method normalizing CT-values to β-actin or gapdh levels.34
Details for primary antibodies used in this investigation are given in Table 2, biotinylated secondary antibodies (antirabbit, antimouse, antigoat) were obtained from Vector Laboratories (Burlingame, CA). For immunohistochemistry a strategy was applied in which the results obtained with a given primary antibody were confirmed by staining the same tissue with an additional antibody obtained from a separate source. Results obtained by this approach revealed consistent staining results. Frozen section (5 μm) were cut and mounted on SuperFrost plus slides (Menzel-Gläser, Braunschweig, Germany). Incubation with primary antibodies was performed for 16 hours at 4°C (VEGF-D, 1:100; VEGFR-3, 1:50; LYVE-1, 1:100; CD-31, 1:100). Negative control stainings performed without addition of primary antibodies did not show immunohistochemical signals and thereby confirmed the specificity of results obtained in stainings with addition of primary antibodies. Following washes in phosphate-buffered saline (PBS), sections were incubated with secondary antibodies (biotinylated goat antirabbit, horse antimouse or rabbit antigoat, respectively, 1:500) and avidin-biotin complex solution (Vectastain Elite ABC Kit 1:250). Stainings were visualized by addition of developing solution (0.2 mL of 38 mmol/L 3-amino-9-ethylcarbazole in 3.8 mL 50 mmol/L acetic buffer pH 5.0). After counterstaining with Hemalaun solution, tissue samples were scored independently by two expert pathologists (S.J., M.V.), who were blinded for clinical data. Staining results for VEGF-C and VEGF-D were classified by estimating the percentage of epithelial cells showing specific immunoreactivity: negative (no immunoreactivity), weak (0% to 5% positive cells), moderate (5% to 50% positive cells), strong (> 50% positive cells). Only samples showing moderate or strong immunoreactivity were considered positive.17 CD31, LYVE-1 or VEGFR-3–positive vessels were determined as described earlier.36 In brief, consecutive sections were stained for CD31, LYVE-1 or VEGFR-3 and scanned for areas with highest vascular densities (hot spots). Vessels of three hot spot areas were counted at a 400-fold magnification in a field area of 0.25 mm2. Staining was considered positive when ≥ 5% of endothelium showed a staining comparable to positive controls included in every run. Correlation of immunohistochemical results with clinicopathologic parameters was performed for an exploratory purpose.
A VEGF-D fragment (Table 1) was selected, subcloned into vector pBluescript II SKH (Stratagene, La Jolla, CA) and sequence confirmed. The probe used for human VEGFR-3 detection has been described before.37 BLAST (basic local alignment search tool; version 2.2.8-2.2.9, National Center for Biotechnology Information, Bethesda, MD) search of cDNA sequences used for in situ hybridization revealed no significant homologies to other genes. After linearization, digoxigenin (DIG)-labeled antisense and sense RNA transcripts were generated by the use of T7 and T3 RNA polymerase, respectively. Tissue cryostat sections (5 μm) were mounted on histoslides, postfixed in 4% paraformaldehyde-PBS (1 hour at 4°C), dehydrated with ethanol and dried. Sections were prehybridized with 100 μL hybridization solution containing 4× saline-sodium citrate, 5% dextransulfate, 1× Denhardt’s solution, 50% formamide, 1 mg/mL yeast tRNA and 1 mg/mL salmon sperm DNA, for 1 hour at 57°C. After incubation with sense or antisense probes, sections were washed and hybridization was visualized by addition of anti-DIG alkaline phosphatase conjugate and nitroblue tetrazolium/5-bromo-4-chloro-3-indiolyl phosphate.
Statistical calculations were performed using the Statistical Package for the Social Sciences software (version 11.0; SPSS Inc., Chicago, IL). Survival was determined from the date of surgery to the time of event (diagnosis of recurrence or death) using the Kaplan-Meier method. Following an intent-to-treat approach, non–cancer-related deaths as well as patients lost to follow-up were included into survival analyses, but censored at the time of death or of last contact, respectively. Relationships between positivity for VEGF-C, VEGF-D and/or VEGFR-3 and clinicopathologic features was evaluated using Spearman’s rank correlation coefficient (ordinally scaled parameters) or Fisher’s exact probability test (dichotome parameters). Statistical significance of differences in cumulative survival curves was evaluated using the log-rank test. In addition to VEGF-C, VEGF-D and VEGFR-3 parameters showing a P value less than .05 in univariate analyses were included in multivariate survival analyses using the Cox proportional hazard method.
Reverse transcriptase polymerase chain reaction determinations revealed that VEGF-C mRNA levels in gastric epithelial cell lines were expressed at low levels (MKN28, CT = 28) or did not differ from background (CT > 35), whereas in endothelial cell line HMEC-1 and HUVECs substantial VEGF-C mRNA levels were detected (CT = 21-22; Fig 1A). In contrast to VEGF-C, VEGF-D gene transcripts were found in gastric cancer cell lines MKN28, KATOIII, and AGS at levels comparable to HMEC-1 cells and HUVECs (CT = 27-30), whereas VEGF-D mRNA levels in MKN45 cells were approximately six- to eight-fold higher (Fig 1B). In keeping with its role as endothelial-specific tyrosine kinase membrane receptor, VEGFR-3 mRNA was not detected in epithelial cancer cell lines (CT > 37) but in endothelial HMEC-1 cells and HUVECs (CT = 27-28; Fig 1C).
To determine the abundance of VEGF-C, VEGF-D, and VEGFR-3 in healthy gastric mucosa, 40 samples of nontumorous healthy mucosa were immunohistochemically stained. Stainings revealed a lack of presence of VEGF-C in healthy gastric mucosa, whereas VEGF-D and VEGFR-3 were found in 12.5% and 83.3% of cases, respectively (data not shown). In contrast to nontumorous gastric mucosa, gastric adenocarcinomas (n = 91) displayed significantly increased abundance of VEGF-C (50.5%; P < .001) or VEGF-D (67.0%; P < .001). In tumors positive for VEGF-C and VEGF-D, both factors were predominantly found in epithelial cells and to a much lesser extent in endothelial cells and/or stromal cells, which, based on histomorphologic criteria, represented fibroblasts, macrophages and/or granulocytes (Fig 2). Presence of VEGF-D in carcinomas tended to be more frequent compared with VEGF-C, without showing a significant difference between proximal and distal gastric cancers (VEGF-D, P = .64; VEGF-C, P = .27) or intestinal and diffuse-type cancers (Table 3). In most of the cancer tissues, VEGF-C and VEGF-D positivity showed a focal distribution pattern, whereas homogenous staining of tumors was found in less than 15% of cases. Results of VEGF-D (Fig 2F) and VEGFR-3 (Fig 2I) in situ hybridizations were similar to corresponding immunohistochemical stainings. Explorative analysis of factors in lymph node metastases (n = 10) revealed that, compared with VEGF-D (80%), VEGF-C (30%) was found less frequently. In primary tumors and metastases, VEGF-C and VEGF-D positivity were not significantly correlated to each other (P = .08). Immunoreactivity of tumors for VEGF-C (P = .006) and VEGF-D (P = .005) was significantly correlated with lymph node metastasis (Table 4). Moreover, VEGF-D, but not VEGF-C, staining was found to be correlated with lymphatic invasion (P = .044) and age ≥ 60 years (P = .020; Table 4). Neither VEGF-C nor VEGF-D positivity was correlated with venous invasion, tumor differentiation, tumor infiltration, patients’ sex and/or tumor localization.
To clarify the relevance of marker positivity for prediction of disease outcome, staining results for VEGF-C, VEGF-D, and VEGFR-3 were correlated with patient survival according to the Kaplan-Meier algorithm. These survival analyses revealed that VEGF-C staining was associated with a nonsignificant trend towards shorter carcinoma-specific survival (P = .114; Fig 3A), whereas a significant correlation with reduced disease-free survival was obtained (P < .05; Fig 3B). In contrast, VEGF-D positivity was significantly correlated with shortened carcinoma-specific (P < .05; Fig 3C) as well as disease-free survival (P < .05; Fig 3D). Moreover, gastric cancers simultaneously positive for VEGF-C and VEGF-D displayed significantly shorter carcinoma-specific (P < .01; Fig 3E) as well as disease-free survival (P < .01; Fig 3F). Multivariate regression analysis revealed that VEGF-D, but not VEGF-C, is an independent prognostic factor for both carcinoma-specific (P = .017) and disease-free survival (P = .023) in gastric adenocarcinomas (Table 5).
In line with previous findings, in our cohort presence of lymph node metastasis and venous tumor invasion, but not lymphatic invasion were identified as independent prognostic factors indicating reduced survival (Table 5).
When compared directly with lymph node positivity (P = .0001; data not shown), which represents the probably most robust prognostic marker in gastric adenocarcinomas, simultaneous tumor positivity for VEGF-C and VEGF-D (P = .006; Fig 3E) did not yield superior prognostic significance in Kaplan and Meier survival analyses (data not shown); however, when analyzed among cases with lymph node metastases (pN+), positivity for VEGF-D (Fig 5B), but not VEGF-C (Fig 5A) or VEGFR-3 (Fig 5C), defined a subset of patients with significantly shorter survival. Moreover, while pN+ cancers positive for VEGF-D and/or VEGFR-3 were associated with strongly reduced 5-year survival (20% to 25%; Fig 5D), absence of VEGFR-3/VEGF-D defined a subgroup of patients displaying a 5-year survival of 100%. The VEGFR-3/VEGF-D-negative pN+ subgroup consisted of cancers classified as pT2 or pT3, which showed except of one pN2 case a pN1 nodal status. Because pN1 gastric cancers are usually associated to a 5-year survival of approximately 40% to 70%,38,39 our results suggest that assessment of VEGF-D/VEGFR-3 tissue status in pN+ gastric cancers provides a novel approach to discriminate patient subgroups with significant differences in their cancer-specific survival and thereby may be useful to facilitate risk-adapted therapy and/or surveillance strategies in these individuals.
VEGFR-3–positive vessels were detected in a subset of gastric cancers as well as samples of healthy gastric mucosa (Table 4; Figs 2 and 4). Most of the vessels showing VEGFR-3 positivity were of thin-walled appearance and did not contain RBCs (Fig 4, top panel, right). In contrast, CD31-positive vessels frequently showed a lumen occasionally containing RBCs, and were characterized by a thicker vascular wall (Fig 4, top panel, left). A large number of VEGFR-3–positive vessels was positive for lymphatic marker LYVE-1, but negative for blood vessel marker CD31 on consecutive sections (Fig 4, middle panel, asterisks). Another subset of VEGFR-3-positive vessels showed strong CD31 staining, demonstrating that VEGFR-3 was also expressed by blood vessel endothelium of gastric cancer tissues (Fig 4, bottom panel, arrowheads and arrows). Moreover, some of the VEGFR-3/CD31-positive vessels were also positive for LYVE-1 (Fig 4, top panel, upper arrows; middle panel, arrowheads), indicating that a subpopulation of VEGFR-3–positive tumor vessels in gastric cancer display features of both, lymphatic and blood vessel endothelial cells. Occasionally, invasion of carcinoma cells into VEGFR-3–positive vessels was observed (not shown). Endothelial VEGFR-3 positivity was significantly correlated with the presence of LYVE-1 (P < .001), but not endothelial CD31 positivity (P = .652) and abundance of VEGF-C or VEGF-D (Table 6) were not. Compared with healthy gastric mucosa, tumor tissues displayed a significant increase in CD-31 as well as LYVE-1 and VEGFR-3–positve vessels (Fig 4A-4C). In an overall analysis, we observed no statistically significant correlations for VEGFR-3 positivity with any of the clinicopathologic parameters analyzed, including lymphatic metastases (Table 4). A subanalysis of gastric cancers with extensive lymphatic spread (N3), however, showed a trend towards correlation between presence of VEGFR-3 and lymphatic metastases (P = .11). Additionally, presence of VEGFR-3–positive vessels clearly correlated with reduced carcinoma-specific survival (P < .05) and showed a trend for correlation with disease-free survival (P = .107; Fig 6). Moreover, cases positive for VEGF-D and VEGFR-3 showed a shorter carcinoma-specific survival compared with cases only positive for VEGF-D (Fig 6C). Similar results were found for disease-free survival; however, differences between individual case groups were less pronounced (Fig 6D). When analyzed by multivariate regression analysis, VEGFR-3 positivity qualified as independent prognostic factor for both carcinoma-specific (P = .016) and disease-free survival (P = .033) showing relative risks of 2.36 and 2.07, respectively, compared with VEGFR-3–negative cases (Table 5).
Here, we report for the first time that presence of VEGF-D correlates with lymphatic metastases, reduced patient survival, and poor prognosis after curative (R0) resection of gastric adenocarcinomas. These results suggest that VEGF-D is linked to lymphatic dissemination and progression of gastric cancer, and indicate that VEGF-D positivity defines a patient subgroup with poor disease outcome. Our results are in agreement with findings in colorectal,15 breast,21 ovarian18 and endometrial carcinomas17 showing that presence of VEGF-D is an independent predictor of poor survival. The correlation between VEGF-D, lymphatic tumor spread, and poor prognosis is in agreement with the established clinical observation that lymphatic dissemination is closely related to the clinical outcome of patients with gastric adenocarcinomas.2,3 However, the observation that in pN+ gastric cancers, presence of VEGF-D is associated with significantly reduced patient survival suggests that mechanisms other than stimulation of nodal dissemination contribute substantially to the survival-relevant pathobiologic effects of VEGF-D in gastric cancer (Fig 5B). Moreover, absence of VEGF-D/VEGFR-3 allows to identify pN+ patients showing favorable 5-year survival, who may, therefore, require specific therapeutic and/or surveillance consideration (Fig 5D). These results show, in addition to their pathobiologic implications, that determination of VEGF-D and/or VEGFR-3 in pN+ gastric cancers allows identification of individuals with significant differences regarding their clinical outcome, and therefore adds to existing diagnostic systems in gastric cancer. Although these observations need further confirmation, assessment of VEGF-D/VEGFR-3 in pN+ patients promises to represent a useful approach establishing risk-adapted therapeutic and/or surveillance strategies in these individuals.
In addition to its role in lymphangiogenesis and lymphatic cancer spread, VEGF-D has been shown to stimulate tumor angiogenesis and progression through interaction with VEGFR-2 and VEGFR-3 on blood vessel endothelial cells.7-10 In our patient cohort, presence of VEGFR-3 was detected on both blood vessel and lymphatic endothelial cells in gastric adenocarcinomas (Fig 4). Moreover, previous work clearly linked VEGFR-2 to angiogenesis and progression of gastric cancer.40,41 Therefore, our results are in line with the concept that in addition to influencing lymphangiogenesis and nodal dissemination, VEGFR-2/VEGFR-3–mediated stimulation of tumor hemangiogenesis may allow VEGF-D to influence the clinical outcome of gastric cancer. A positive correlation between VEGF-C expression and lymphatic invasion,26-29 lymph node metastasis27-29 and/or venous invasion29 has been described by several groups in gastric and nongastric cancers (see review in Duff et al42); however, the association between VEGF-C and patients’ prognosis remained less well understood. No significant relationship between VEGF-C and survival was found in gastric,26 esophageal,43 and colorectal cancers,44 whereas others found that high levels of VEGF-C were associated with poor prognosis and decreased survival.27,29 In our study, VEGF-C positivity was associated with presence of lymphatic metastases (P < .01; Table 4) and reduced disease-free survival (P < .05) (Fig 3B). In contrast to VEGF-D, VEGF-C failed to achieve status as independent prognostic factor in multivariate Cox regression analysis. Remarkably, patients with tumors simultaneously positive for VEGF-C and VEGF-D showed a clearly shorter survival compared with cancers that were positive for either VEGF-C or VEGF-D (Fig 3E and 3F).
Similar to our observations, differences in the pathobiologic profiles of VEGF-C and VEGF-D have been observed by others: Overexpression of VEGF-C in a transgenic mouse model resulted in increased lymphatic metastases, whereas tumor volume and/or tumor angiogenesis were not influenced.45 Similarly, murine xenograft tumors of human breast cancer cells overexpressing VEGF-C showed increased lymphangiogenesis, whereas tumor angiogenesis was not changed.46,47 In contrast, overexpression of VEGF-D clearly increased lymphangiogenesis and lymphatic metastases as well as tumor growth and angiogenesis.48 It has been confirmed that progressive proteolytic processing determines the affinity of VEGF-C and VEGF-D towards VEGFR-2, and thereby influences the functional profiles elicited by both ligands.12 In line with this concept, the VEGFR-3–preferring 31-kD VEGF-C isoform was found to stimulate lymphangiogenesis and lymphatic metastases in mouse models,46,49 whereas expression of the 21-kD VEGF-C isoform, capable of activating both VEGFR-2 and VEGFR-3 receptors,12 increased both, lymphangiogenesis and angiogenesis.7,50 As yet it is unclear which molecular forms of VEGF-C and VEGF-D are present in gastric cancer tissues and whether differences in the molecular compositions of VEGF-C or VEGF-D isoforms may be related to different clinical outcomes. Moreover, the system of proteases responsible for cleavage of VEGF peptides is only understood preliminarily and needs further clarification.13 Nevertheless, characterization of VEGF-C/VEGF-D isoforms and related proteases in relation to clinical disease outcomes may further extend the diagnostic and/or prognostic value of these factors in clinical oncology.
In a recent study, the abundance of VEGF-C and VEGF-D in early gastric carcinomas was investigated.51 Similar to our results in advanced gastric adenocarcinomas, presence of VEGF-C and VEGF-D was found to correlate with nodal status and lymphatic invasion. However, in contrast to our observations and findings of other investigators in gastric26-29,52 and nongastric cancers,15,17-19 Ishikawa et al51 reported an inverse relationship between VEGF-C and VEGF-D expression and the grade of tumor differentiation. This finding is in clear contradiction to the common clinical observation that undifferentiated gastric adenocarcinomas are frequently associated with lymph node metastases and therefore requires further confirmation. Moreover, Ishikawa et al did not, unfortunately, investigate the correlation between VEGF-C/VEGF-D tissue status and patient prognosis. Therefore, it remains to be answered whether analysis of VEGF-C/VEGF-D in early gastric cancer may also be of prognostic value as observed by us after curative resection of advanced gastric cancer.
In contrast to the clearly confirmed importance of VEGFR-3 in endothelial cell biology, results on the correlation of VEGFR-3 with clinicopathologic parameters and the prognosis of cancer patients are heterogenous. In breast,22,23 endometrial,17 colorectal53 and non–small-cell lung cancer20 as well as squamous cancer of the tongue,54 endothelial expression of VEGFR-3 did not correlate with lymphatic metastases. In contrast, a positive correlation between VEGFR-3 and nodal status was found in oral squamous carcinoma55 and breast cancer.21 In gastric adenocarcinomas, the number of VEGFR-3–positive vessels correlated with nodal status and lymphatic invasion,30 whereas another study by the same group did not find such correlation when VEGFR-3 RNA was analyzed.29 In our patient cohort, VEGFR-3 immunopositivity was associated significantly with reduced carcinoma-specific survival (Fig 6); similar to results in non–small-cell lung cancer20 and endometrial carcinoma,17 VEGFR-3 qualified as independent prognostic marker for reduced survival (Table 5). Expression of VEGFR-3 in cancer tissues has been found to correlate with the presence of its ligands VEGF-C and VEGF-D in cancer tissues.17,18,20,21,29,30 In contrast to these studies and in agreement with observations made by White et al15 in colorectal cancer samples, we did not find a correlation between VEGFR-3 and expression of its ligands. However, in our patient cohort, VEGFR-3 positivity was closely related to the presence of the lymphatic marker LYVE-1 (P < .001; Table 6), and VEGFR-3–positive tumors showed an increased density of VEGFR-3/LYVE-1-positive vessels representing tumor lymphatics (Fig 4). Although these results suggest a close relationship between VEGFR-3 and lymphatic tumor vascularization in gastric adenocarcinoma, a correlation between VEGFR-3 and lymphatic metastases was found only in advanced gastric cancers (N3) displaying extensive lymphatic dissemination, however, without reaching statistical significance (P = .11). This result is in agreement with the observation that in contrast to preclinical murine tumor models,45,56,57 clinical studies analyzing patient samples often failed to show a correlation between VEGFR-3 expression and clinical end points such as lymphatic dissemination,17,20,22,23,53,54 questioning the requirement of increased lymphatic vascularization for lymphatic tumor spread (see review in Jain and Fenton58). As a potential explanation for this phenomenon, a recent study showed that lymphatic metastases can occur in the absence of functional intratumoral lymphatics and suggested that an increase in the surface area of lymphatics located at the tumor margin may serve as possible underlying mechanism.59 To what extent an increased number of intratumoral lymphatic vessels and/or alternative mechanisms like an increase in the surface area of peripheral tumor lymphatics may contribute to the nodal dissemination of solid tumors including gastric cancer is currently unclear and requires further clarification.58
In conclusion, we identify VEGF-D and VEGFR-3 as novel prognostic marker molecules for reduced survival after curative resection of gastric adenocarcinomas. Our results show that analysis of the VEGF-C/VEGF-D/VEGFR-3 system in gastric cancer tissues can help to identify patient subgroups at higher risk for poor disease outcome, who therefore may deserve specific therapeutic consideration. Finally, our findings suggest that interference with VEGF-C/VEGF-D/VEGFR-3–dependent pathways may represent an attractive, innovative therapeutic concept for the treatment of gastric cancer patients.
The authors indicated no potential conflicts of interest.
Conception and design: Stefan Jüttner, Stephan Gretschel, Wolfgang Kemmner, Michael Höcker
Financial support: Michael Höcker
Administrative support: Peter M. Schlag, Wolfgang Kemmner, Michael Höcker
Provision of study materials or patients: Stephan Gretschel, Peter M. Schlag, Wolfgang Kemmner
Collection and assembly of data: Stefan Jüttner, Christoph Wiβmann, Thomas Jöns, Johannes Hertel
Data analysis and interpretation: Stefan Jüttner, Christoph Wiβmann, Thomas Jöns, Michael Vieth, Michael Höcker
Manuscript writing: Stefan Jüttner, Christoph Wiβmann, Michael Höcker
Final approval of manuscript: Stefan Jüttner, Christoph Wiβmann, Thomas Jöns, Michael Vieth, Stephan Gretschel, Peter M. Schlag, Wolfgang Kemmner, Michael Höcker
| Angiogenesis: | The process involved in the generation of new blood vessels. While this is a normal process that naturally occurs and is controlled by “on” and “of” switches, blocking tumor angiogenesis (antiangiogenesis) disrupts the blood supply to tumors, thereby preventing tumor growth. |
| Carcinoma-specific survival: | Survival period spanning the time from surgery to death due to cancer. |
| Disease-free survival: | Survival period spanning the time from surgery to recurrence of cancer. |
| Lymphangiogenesis: | Formation of the lymphatic network of capillaries. Unlike capillaries of the blood vascular system, lymphatic capillaries are characterized by a lining of a single layer of endothelial cells, devoid of fenestrations, with poorly developed junctions and the presence of frequent, large interendothelial gaps. Additionally, lymphatic capillaries lack a continuous basement membrane and are devoid of pericytes. Although lymphangiogenesis has an important physiological role in homeostasis, metabolism, and immunity, it has also been implicated in diseases such as neoplasm metastasis, edema, rheumatoid arthritis, psoriasis, and impaired wound healing. Podoplanin, LYVE- (lymphatic vessel endothelial hyaluronan receptor) 1, PROX-1, desmoplakin, and the VEGF-C/VEGF-D receptor VEGFR-3 are important markers specific to lymphangiogenesis. |
| Lymphatic metastases: | Tumor spread via lymphatics affecting lymph nodes. Lymphatic metastases are linked to lymphatic dissemination of tumors, which occurs either via intratumoral lymphatics or due to an increase in surface area of lymphatics at tumor margins. Vascular endothelial growth factor (VEGF) -C and VEGF-D are growth factors linked to lymphatic dissemination of tumors and, consequently, along with VEGFR-3 are important predictors of poor survival. |
| Vasculogenesis: | Early development of the vascular system whereby new capillaries are formed. According to two separate proposals, the mechanisms that explain vasculogenesis are related to the sprouting of new capillaries from newly formed vessels (capillaries or veins) or new capillaries arising de novo from mesoderm-derived angioblasts. |
| VEGF (vascular endothelial growth factor): | VEGF is a cytokine that mediates numerous functions of endothelial cells including proliferation, migration, invasion, survival, and permeability. VEGF is also known as vascular permeability factor. VEGF naturally occurs as a glycoprotein and is critical for angiogenesis. Many tumors overexpress VEGF, which correlates to poor prognosis. VEGF-A, -B, -C, -D, and -E are members of the larger family of VEGF-related proteins. |
| VEGFR (vascular endothelial growth factor receptor): | VEGFRs are transmembrane tyrosine kinase receptors to which the VEGF ligand binds. VEGFR-1 (also called Flt-1) and VEGFR-2 (also called KDR/Flk-1[murine homologue]) are expressed on endothelial cells, while VEGFR-3 (also called Flt-4) is expressed on cells of the lymphatic and vascular endothelium. VEGFR-2 is thought to be principally responsible for angiogenesis and for the proliferation of endothelial cells. Typically, most VEGFRs have seven extracellular immunoglobulin-like domains, responsible for VEGF binding, and an intracellular tyrosine kinase domain. |
Fig 1. Real time reverse transcriptase polymerase chain reaction determination of (A) vascular endothelial growth factor (VEGF) -C, (B) VEGF-D, AND (C) VEGFR-3 mRNA levels in gastric cancer cell lines MKN45, MKN28, KatoIII and AGS. As controls, mRNAs from primary human umbilical vein endothelial cells (HUVECs) and permanent endothelial cell line HMEC-1 were analyzed. The data shown represent typical results obtained in a series of three independent experiments.
Fig 2. Distribution of vascular endothelial growth factor (VEGF) -C, VEGF-D and VEGFR-3 in gastric adenocarcinomas and lymph node metastasis. Typical examples of VEGF-C immunopositivity in gastric adenocarcinomas at (A) ×200 and (B) ×400; (C) VEGF-C–positive cancer cells in a lymph node metastasis (×100). Typical results of VEGF-D immunopositivity in gastric cancers at (D) ×100 and (E) ×100; (F) a representative result for VEGF-D in situ hybridization at ×100 (identical tumor as shown in E; the inset shows the sense-control). Inset in panels B and E show staining results with anti-VEGF-C or anti-VEGF-D antibodies, respectively, in tumors negative for the respective factor. Typical patterns of VEGF receptor (VEGFR) -3 immunopositivity in gastric adenocarcinomas at (G) ×100 and (H) ×100. Endothelial cells of different vessels as well as tumor epithelial cells were immunodecorated (Fig 4). (I) Results obtained with in situ hybridization using a VEGFR-3–specific riboprobe confirm the results obtained with immunohistochemistry (×100; inset shows sense-control of the same tumor).
Fig 3. The relationship between vascular endothelial growth factor (VEGF) -C (A, B), VEGF-D (C, D) or simultaneous immunopositivity for both factors (E, F) and survival of gastric cancer patients was assessed employing the Kaplan-Meier method. Panels on the left show results for carcinoma-specific survival, whereas panels on the right represent data for disease-free survival. Statistical evaluation was performed by using the log-rank test.
Fig 4. Representative results for (A) CD31, (B) LYVE-1, and (C) vascular endothelial growth factor receptor (VEGFR) -3 staining of consecutive sections of gastric adenocarcinomas (×100); highlighted squares mark the sections that are shown at higher magnification in the middle and bottom panel (×400). A large number of VEGFR-3–positive vessels was positive for lymphatic marker LYVE-1, but did not show staining for blood vessel marker CD31 (middle panels, asterisks). Another subset of VEGFR-3–positive vessels showed strong CD31 staining, demonstrating that VEGFR-3 was also expressed by blood vessel endothelium of gastric cancer tissues (bottom panels, arrowheads and arrows). Some of the VEGFR-3/CD31–positive vessels were also positive for LYVE-1 (top panel, arrows; middle panel, arrowheads). The number of (D) CD31, (E) LYVE-1, and (F) VEGFR-3-positive vessels in healthy gastric mucosa and gastric cancer samples. Statistical significances between mean vessel counts were evaluated using Student’s t test. (**), P < .01; (***), P < .001).
Fig 6. The relationship between presence of vascular endothelial growth factor receptor (VEGFR)-3 immunopositivity and (A) carcinoma-specific as well as (B) disease-free survival was assessed employing the Kaplan-Meier method. In addition, cumulative survival of patients showing simultaneous positivity for vascular endothelial growth factor receptor (VEGFR)-3 and VEGF-D was determined and compared with patients who were positive for either VEGFR-3 or VEGF-D (C, D). Statistical evaluation was performed by using the log-rank test.
Fig 5. Relationships between immunopositivity for (A) vascular endothelial growth factor (VEGF)-C, (B) VEGF-D, or (C) VEGFR-3 and carcinoma-specific survival were assessed in pN+ gastric adenocarcinomas employing the Kaplan-Meier method. (D) Simultaneous immunopositivity for VEGF-D and VEGFR-3. Statistical evaluation was performed using the log-rank test.
|
| Forward (5′ → 3′) | Reverse (5′ → 3′) | Probe (5′-Fam → 3′-Tamra) | |
|---|---|---|---|
| VEGF-C | GTTCCACCACCAAACATGCA | CACTATATGAAAATCCTGGCTCACAA | ACGGCCATGTACGAACCGCCA |
| VEGF-D | CTGGAACAGAAGACCACTCTCATC | CTCGCAACGATCTTCGTCAAA | CAGGAACCAGCTCTCTGTGGGC |
| VEGFR-3 | GCTGAGACCCGTGGTTCCT | CTATGCCTGCTCTCTATCTGCTCAA | ACGACCTACAAAGGCTCTGTGGACAACCA |
| GAPDH | GAAGGTGAAGGTCGGAGTC | GAAGATGGTGATGGGATTTC | CAAGCTTCCCGTTCTCAGCC |
| β-ACTIN | TGCATTGTTACAGGAAGTCCCTT | GGGAGAGGACTGGGCCAT | CCATCCTAAAAGCCACCCCACTTCTCTCTA |
| VEGF-D (in situ) | ATCGGGTACCTTTGCAGGAGGAAAATCCAC | ATCGGGATCCTCTGGTATGAAAGGGGCATC |
Abbreviations: Fam, 6-carboxyfluorescein; Tamra, 6-carboxytetramethylrhodamine; VEGF, vascular endothelial growth factor; VEGFR, VEGF receptor.
|
| Antigen | Antibody | No. |
|---|---|---|
| VEGF-C | Polyclonal rabbit (Zymed Laboratories Inc, Carlsbad, CA) | 18-2255 |
| VEGF-C | Polyclonal goat (Santa Cruz Biotechnology Inc, Santa Cruz, CA) | (N-19) sc-7133 |
| VEGF-D | Polyclonal goat (Santa Cruz Biotechnology) | (C-18) sc-7602 |
| VEGF-D | Polyclonal goat (R&D Systems Inc, Minneapolis, MN) | AF286 |
| VEGFR-3 | Polyclonal rabbit (RELIATech GmbH, Braunschweig, Germany) | 102-PA22 |
| VEGFR-3 | Polyclonal rabbit (Santa Cruz Biotechnology) | (C-20) sc-321 |
| LYVE-1 | Polyclonal rabbit (RELIATech GmbH) | 102-PA50S |
| LYVE-1 | Polyclonal rabbit (D. Jackson, Oxford) | Banerji et al35 |
| CD31 | Monoclonal mouse (Dako Cytomation, Carpinteria, CA) | M 0823 |
Abbreviations: VEGF, vascular endothelial growth factor; VEGFR, VEGF receptor.
|
| No. of Cases | Intestinal Type | No. of Cases | Diffuse Type | P | |||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| + | − | + | − | ||||||||||||||
| No. | % | No. | % | No. | % | No. | % | ||||||||||
| VEGF-C | 48 | 26 | 54.2 | 22 | 45.8 | 40 | 18 | 45.0 | 22 | 55.0 | .521 | ||||||
| Total | 24 | 13 | 54.2 | 11 | 45.8 | 10 | 7 | 70.0 | 3 | 30.0 | .467 | ||||||
| Proximal distal | 24 | 13 | 54.2 | 11 | 45.8 | 30 | 11 | 36.7 | 19 | 63.3 | .272 | ||||||
| VEGF-D | 48 | 32 | 66.7 | 16 | 33.3 | 40 | 27 | 67.5 | 13 | 32.5 | .99 | ||||||
| Total | 24 | 15 | 62.5 | 9 | 37.5 | 10 | 9 | 90.0 | 1 | 10.0 | .215 | ||||||
| Proximal distal | 24 | 17 | 70.8 | 7 | 29.2 | 30 | 18 | 60.0 | 12 | 40.0 | .567 | ||||||
| VEGFR-3 | 47 | 15 | 31.9 | 32 | 68.1 | 38 | 14 | 36.8 | 24 | 63.2 | .653 | ||||||
| Total | 23 | 4 | 17.4 | 19 | 82.6 | 10 | 1 | 10.0 | 9 | 90.0 | .99 | ||||||
| Proximal distal | 24 | 11 | 45.8 | 13 | 54.2 | 28 | 13 | 46.4 | 15 | 53.6 | .99 | ||||||
NOTE. Correlations were calculated using Fisher’s exact test.
Abbreviations: VEGF, vascular endothelial growth factor; VEGFR, VEGF receptor.
|
| No. of Cases | VEGF-C | VEGF-D | No. of Cases | VEGFR-3 | |||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| No. | % | P | No. | % | P | No. | % | P | |||||||||
| Lymphatic invasion | .095 | .044 | .498 | ||||||||||||||
| Absent | 46 | 19 | 41.3 | 26 | 56.5 | 45 | 13 | 28.9 | |||||||||
| Present | 45 | 27 | 60.0 | 35 | 77.8 | 43 | 16 | 37.2 | |||||||||
| Venous invasion | .640 | .99 | .620 | ||||||||||||||
| Absent | 66 | 32 | 48.5 | 44 | 66.7 | 63 | 22 | 34.9 | |||||||||
| Present | 25 | 14 | 56.0 | 17 | 68.0 | 25 | 7 | 28.0 | |||||||||
| Lymph node metastasis | .006 | .005 | .288 | ||||||||||||||
| N0 | 28 | 8 | 28.6 | 16 | 57.1 | 26 | 7 | 26.9 | |||||||||
| N1 | 31 | 19 | 61.3 | 16 | 51.6 | 32 | 11 | 34.4 | |||||||||
| N2 | 21 | 10 | 47.6 | 19 | 90.5 | 20 | 5 | 25.0 | |||||||||
| N3 | 11 | 9 | 81.8 | 10 | 90.9 | 10 | 6 | 60.0 | |||||||||
| Tumor differentiation | .246 | .171 | .787 | ||||||||||||||
| G1 | 2 | 0 | 2 | 100 | 2 | 0 | |||||||||||
| G2 | 19 | 7 | 36.8 | 14 | 73.7 | 19 | 6 | 31.6 | |||||||||
| G3 | 67 | 39 | 58.2 | 44 | 65.7 | 65 | 23 | 35.4 | |||||||||
| G4 | 3 | 0 | 1 | 33.3 | 2 | 0 | |||||||||||
| Tumor infiltration | .168 | .871 | .869 | ||||||||||||||
| T1 | 14 | 5 | 35.7 | 9 | 64.3 | 13 | 4 | 30.8 | |||||||||
| T2 | 50 | 25 | 50.0 | 34 | 68.0 | 48 | 16 | 33.3 | |||||||||
| T3 | 22 | 13 | 59.1 | 14 | 63.6 | 22 | 7 | 31.8 | |||||||||
| T4 | 5 | 3 | 60.0 | 4 | 80 | 5 | 2 | 40 | |||||||||
| Age, years | .660 | .020 | .052 | ||||||||||||||
| < 60 | 30 | 14 | 46.7 | 15 | 50.0 | 29 | 14 | 48.3 | |||||||||
| ≥ 60 | 61 | 32 | 52.5 | 46 | 75.4 | 59 | 15 | 25.4 | |||||||||
| Sex | .512 | .99 | .99 | ||||||||||||||
| Male | 59 | 28 | 46.5 | 39 | 66.1 | 58 | 19 | 32.8 | |||||||||
| Female | 32 | 18 | 56.3 | 22 | 68.8 | 30 | 10 | 33.3 | |||||||||
NOTE. Boldfacing indicates statistically significant correlations. Results were calculated by Spearman correlation coefficient or Fisher’s exact test.
Abbreviations: VEGF, vascular endothelial growth factor; VEGFR, VEGF receptor.
|
| Factor | Carcinoma-Specific Survival | Disease-Free Survival | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Relative Risk | 95% CI | P | Relative Risk | 95% CI | P | |||||
| VEGF-C | ||||||||||
| Negative | 1 | 1 | ||||||||
| Positive | 0.86 | 0.42 to 1.75 | .675 | 1.26 | 0.66 to 2.41 | .485 | ||||
| VEGF-D | ||||||||||
| Negative | 1 | 1 | ||||||||
| Positive | 3.08 | 1.22 to 7.80 | .017 | 2.65 | 1.14 to 6.16 | .023 | ||||
| VEGFR-3 | ||||||||||
| Negative | 1 | 1 | ||||||||
| Positive | 2.36 | 1.17 to 4.74 | .016 | 2.07 | 1.06 to 4.04 | .033 | ||||
| Lymphatic invasion | ||||||||||
| Absent | 1 | 1 | ||||||||
| Present | 1.81 | 0.80 to 4.12 | .157 | 1.71 | 0.82 to 3.58 | .155 | ||||
| Venous invasion | ||||||||||
| Absent | 1 | 1 | ||||||||
| Present | 3.30 | 1.55 to 7.02 | .002 | 4.19 | 2.05 to 8.58 | < .001 | ||||
| Lymph node metastasis | ||||||||||
| Absent | 1 | 1 | ||||||||
| Present | 3.13 | 0.97 to 10.13 | .056 | 3.44 | 1.10 to 10.82 | .034 | ||||
NOTE. Boldfacing indicates statistically significant results.
Abbreviations: VEGF, vascular endothelial growth factor; VEGFR, VEGF receptor.
|
| No. of Cases | VEGF-C | VEGF-D | VEGFR-3 | |||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| No. | % | P | No. | % | P | No. | % | P | ||||||||
| VEGFR-3 | .370 | .99 | ||||||||||||||
| Negative | 59 | 28 | 47.5 | 40 | 67.8 | — | — | |||||||||
| Positive | 29 | 17 | 58.6 | 20 | 69.0 | |||||||||||
| LYVE-1 | .832 | .494 | < .0001 | |||||||||||||
| Negative | 52 | 26 | 50.0 | 37 | 71.2 | 6 | 11.5 | |||||||||
| Positive | 36 | 19 | 52.8 | 23 | 63.9 | 23 | 63.9 | |||||||||
| CD-31 | .087 | .108 | .652 | |||||||||||||
| < median | 44 | 27 | 61.4 | 34 | 77.3 | 13 | 29.6 | |||||||||
| ≥ median | 44 | 18 | 40.9 | 26 | 59.1 | 16 | 36.4 | |||||||||
NOTE. Boldfacing indicates statistically significant correlations. Results were calculated by Fisher’s exact test.
Abbreviations: VEGF, vascular endothelial growth factor; VEGFR, VEGF receptor.
Supported by Grant No. NBLIII TP 3.2 from the Bundesministerium für Bildung und Forschung and by a grant from the Sonnenfeld Stiftung, Berlin, Germany.
S.J. and C.W. contributed equally to this manuscript.
Terms in blue are defined in the glossary, found at the end of this article and online at www.jco.org.
Authors’ disclosures of potential conflicts of interest and author contributions are found at the end of this article.
We gratefully acknowledge the excellent technical assistance of Carola Meier.
| 1. | Parkin DM: Global cancer statistics in the year 2000. Lancet Oncol 2::533,2001-543, Crossref, Medline, Google Scholar |
| 2. | Hohenberger P, Gretschel S: Gastric cancer. Lancet 362::305,2003-315, Crossref, Medline, Google Scholar |
| 3. | Maehara Y, Kakeji Y, Koga T, et al: Therapeutic value of lymph node dissection and the clinical outcome for patients with gastric cancer. Surgery 131::S85,2002-S91, Crossref, Medline, Google Scholar |
| 4. | Tunaci M: Carcinoma of stomach and duodenum: Radiologic diagnosis and staging. Eur J Radiol 42::181,2002-192, Crossref, Medline, Google Scholar |
| 5. | Yancopoulos GD, Davis S, Gale NW, et al: Vascular-specific growth factors and blood vessel formation. Nature 407::242,2000-248, Crossref, Medline, Google Scholar |
| 6. | Carmeliet P, Jain RK: Angiogenesis in cancer and other diseases. Nature 407::249,2000-257, Crossref, Medline, Google Scholar |
| 7. | Stacker SA, Achen MG, Jussila L, et al: Lymphangiogenesis and cancer metastasis. Nat Rev Cancer 2::573,2002-583, Crossref, Medline, Google Scholar |
| 8. | Ferrara N, Gerber HP, LeCouter J: The biology of VEGF and its receptors. Nat Med 9::669,2003-676, Crossref, Medline, Google Scholar |
| 9. | Karkkainen MJ, Petrova TV: Vascular endothelial growth factor receptors in the regulation of angiogenesis and lymphangiogenesis. Oncogene 19::5598,2000-5605, Crossref, Medline, Google Scholar |
| 10. | Veikkola T, Karkkainen M, Claesson-Welsh L, et al: Regulation of angiogenesis via vascular endothelial growth factor receptors. Cancer Res 60::203,2000-212, Medline, Google Scholar |
| 11. | Stacker SA, Stenvers K, Caesar C, et al: Biosynthesis of vascular endothelial growth factor-D involves proteolytic processing which generates non-covalent homodimers. J Biol Chem 274::32127,1999-32136, Crossref, Medline, Google Scholar |
| 12. | Joukov V, Sorsa T, Kumar V, et al: Proteolytic processing regulates receptor specificity and activity of VEGF-C. EMBO J 16::3898,1997-3911, Crossref, Medline, Google Scholar |
| 13. | McColl BK, Baldwin ME, Roufail S, et al: Plasmin activates the lymphangiogenic growth factors VEGF-C and VEGF-D. J Exp Med 198::863,2003-868, Crossref, Medline, Google Scholar |
| 14. | Paavonen K, Puolakkainen P, Jussila L, et al: Vascular endothelial growth factor receptor-3 in lymphangiogenesis in wound healing. Am J Pathol 156::1499,2000-1504, Crossref, Medline, Google Scholar |
| 15. | 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 62::1669,2002-1675, Medline, Google Scholar |
| 16. | Onogawa S, Kitadai Y, Tanaka 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 95::32,2004-39, Crossref, Medline, Google Scholar |
| 17. | Yokoyama Y, Charnock-Jones DS, Licence D, et al: Expression of vascular endothelial growth factor (VEGF)-D and its receptor, VEGF receptor 3, as a prognostic factor in endometrial carcinoma. Clin Cancer Res 9::1361,2003-1369, Medline, Google Scholar |
| 18. | Yokoyama Y, Charnock-Jones DS, Licence D, et al: Vascular endothelial growth factor-D is an independent prognostic factor in epithelial ovarian carcinoma. Br J Cancer 88::237,2003-244, Crossref, Medline, Google Scholar |
| 19. | 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 9::716,2003-721, Medline, Google Scholar |
| 20. | Arinaga M, Noguchi T, Takeno S, et al: Clinical significance of vascular endothelial growth factor C and vascular endothelial growth factor receptor 3 in patients with nonsmall cell lung carcinoma. Cancer 97::457,2003-464, Crossref, Medline, Google Scholar |
| 21. | Nakamura Y, Yasuoka H, Tsujimoto M, et al: Flt-4-positive vessel density correlates with vascular endothelial growth factor-d expression, nodal status, and prognosis in breast cancer. Clin Cancer Res 9::5313,2003-5317, Medline, Google Scholar |
| 22. | Gunningham SP, Currie MJ, Han C, et al: 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 6::4278,2000-4286, Medline, Google Scholar |
| 23. | Jacquemier J, Mathoulin-Portier MP, Valtola R, et al: Prognosis of breast-carcinoma lymphagenesis evaluated by immunohistochemical investigation of vascular-endothelial-growth-factor receptor 3. Int J Cancer 89::69,2000-73, Crossref, Medline, Google Scholar |
| 24. | O-charoenrat P, 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 92::556,2001-568, Crossref, Medline, Google Scholar |
| 25. | Niki T, Iba S, Tokunou M, et al: Expression of vascular endothelial growth factors A, B, C, and D and their relationships to lymph node status in lung adenocarcinoma. Clin Cancer Res 6::2431,2000-2439, Medline, Google Scholar |
| 26. | Ichikura T, Tomimatsu S, Ohkura E, et al: Prognostic significance of the expression of vascular endothelial growth factor (VEGF) and VEGF-C in gastric carcinoma. J Surg Oncol 78::132,2001-137, Crossref, Medline, Google Scholar |
| 27. | Takahashi A, Kono K, Itakura J, et al: Correlation of vascular endothelial growth factor-C expression with tumor-infiltrating dendritic cells in gastric cancer. Oncology 62::121,2002-127, Crossref, Medline, Google Scholar |
| 28. | 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 38::1413,2002-1419, Crossref, Medline, Google Scholar |
| 29. | 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 5::1823,1999-1829, Medline, Google Scholar |
| 30. | Yonemura Y, Fushida S, Bando E, et al: Lymphangiogenesis and the vascular endothelial growth factor receptor (VEGFR)-3 in gastric cancer. Eur J Cancer 37::918,2001-923, Crossref, Medline, Google Scholar |
| 31. | Liu XE, Sun XD, Wu JM: Expression and significance of VEGF-C and FLT-4 in gastric cancer. World J Gastroenterol 10::352,2004-355, Crossref, Medline, Google Scholar |
| 32. | Hamilton SR, Aaltonen LA WHO Classification. Tumours of the Digestive System: Pathology & Genetics . Lyon, France, IARC Press, 2000 Google Scholar |
| 33. | Wittekind C, Meyer HJ, Bootz F: TNM Classification of Malignant Tumors (ed 6) . New York, NY, John Wiley & Sons Inc, 2002 Google Scholar |
| 34. | Schäfer G, Cramer T, Suske G, et al: Oxidative stress regulates vascular endothelial growth factor-A gene transcription through Sp1- and Sp3-dependent activation of two proximal GC-rich promoter elements. J Biol Chem 278::8190,2003-8198, Crossref, Medline, Google Scholar |
| 35. | Banerji S, Ni J, Wang SX, et al: LYVE-1, a new homologue of the CD44 glycoprotein, is a lymph-specific receptor for hyaluronan. J Cell Biol 144::789,1999-801, Crossref, Medline, Google Scholar |
| 36. | Weidner N: Tumor angiogenesis: Review of current applications in tumor prognostication. Semin Diagn Pathol 10::302,1993-313, Medline, Google Scholar |
| 37. | Huang HY, Ho CC, Huang PH, et al: Co-expression of VEGF-C and its receptors, VEGFR-2 and VEGFR-3, in endothelial cells of lymphangioma. Implication in autocrine or paracrine regulation of lymphangioma. Lab Invest 81::1729,2001-1734, Crossref, Medline, Google Scholar |
| 38. | Katai H, Yoshimura K, Maruyama K, et al: Evaluation of the New International Union Against Cancer TNM staging for gastric carcinoma. Cancer 88::1796,2000-1800, Crossref, Medline, Google Scholar |
| 39. | Klein KE, Hermans J, van Krieken JH, et al: Evaluation of the 5th edition of the TNM classification for gastric cancer: Improved prognostic value. Br J Cancer 84::64,2001-71, Crossref, Medline, Google Scholar |
| 40. | Jung YD, Mansfield PF, Akagi M, et al: Effects of combination anti-vascular endothelial growth factor receptor and anti-epidermal growth factor receptor therapies on the growth of gastric cancer in a nude mouse model. Eur J Cancer 38::1133,2002-1140, Crossref, Medline, Google Scholar |
| 41. | Kamiyama M, Ichikawa Y, Ishikawa T, et al: VEGF receptor antisense therapy inhibits angiogenesis and peritoneal dissemination of human gastric cancer in nude mice. Cancer Gene Ther 9::197,2002-201, Crossref, Medline, Google Scholar |
| 42. | Duff SE, Li C, Jeziorska M, et al: Vascular endothelial growth factors C and D and lymphangiogenesis in gastrointestinal tract malignancy. Br J Cancer 89::426,2003-430, Crossref, Medline, Google Scholar |
| 43. | Kitadai Y, Amioka T, Haruma K, et al: Clinicopathological significance of vascular endothelial growth factor (VEGF)-C in human esophageal squamous cell carcinomas. Int J Cancer 93::662,2001-666, Crossref, Medline, Google Scholar |
| 44. | Akagi K, Ikeda Y, Miyazaki M, et al: Vascular endothelial growth factor-C (VEGF-C) expression in human colorectal cancer tissues. Br J Cancer 83::887,2000-891, Crossref, Medline, Google Scholar |
| 45. | Mandriota SJ, Jussila L, Jeltsch M, et al: Vascular endothelial growth factor-C-mediated lymphangiogenesis promotes tumour metastasis. EMBO J 20::672,2001-682, Crossref, Medline, Google Scholar |
| 46. | Skobe M, Hawighorst T, Jackson DG, et al: Induction of tumor lymphangiogenesis by VEGF-C promotes breast cancer metastasis. Nat Med 7::192,2001-198, Crossref, Medline, Google Scholar |
| 47. | Karpanen T, Egeblad M, Karkkainen MJ, et al: Vascular endothelial growth factor C promotes tumor lymphangiogenesis and intralymphatic tumor growth. Cancer Res 61::1786,2001-1790, Medline, Google Scholar |
| 48. | Stacker SA, Caesar C, Baldwin ME, et al: VEGF-D promotes the metastatic spread of tumor cells via the lymphatics. Nat Med 7::186,2001-191, Crossref, Medline, Google Scholar |
| 49. | Karpanen T, Alitalo K: Lymphatic vessels as targets of tumor therapy? J Exp Med 194::F37,2001-F42, Crossref, Medline, Google Scholar |
| 50. | Skobe M, Hamberg LM, Hawighorst T, et al: Concurrent induction of lymphangiogenesis, angiogenesis, and macrophage recruitment by vascular endothelial growth factor-C in melanoma. Am J Pathol 159::893,2001-903, Crossref, Medline, Google Scholar |
| 51. | Ishikawa M, Kitayama J, Kazama S, et al: Expression of vascular endothelial growth factor C and D (VEGF-C and -D) is an important risk factor for lymphatic metastasis in undifferentiated early gastric carcinoma. Jpn J Clin Oncol 33::21,2003-27, Crossref, Medline, Google Scholar |
| 52. | Kabashima A, Maehara Y, Kakeji Y, et al: Overexpression of vascular endothelial growth factor C is related to lymphogenous metastasis in early gastric carcinoma. Oncology 60::146,2001-150, Crossref, Medline, Google Scholar |
| 53. | George ML, Tutton MG, Janssen F, et al: VEGF-A, VEGF-C, and VEGF-D in colorectal cancer progression. Neoplasia 3::420,2001-427, Crossref, Medline, Google Scholar |
| 54. | Okamoto M, Nishimine M, Kishi M, et al: Prediction of delayed neck metastasis in patients with stage I/II squamous cell carcinoma of the tongue. J Oral Pathol Med 31::227,2002-233, Crossref, Medline, Google Scholar |
| 55. | Moriyama M, Kumagai S, Kawashiri S, et al: Immunohistochemical study of tumour angiogenesis in oral squamous cell carcinoma. Oral Oncol 33::369,1997-374, Crossref, Medline, Google Scholar |
| 56. | He Y, Kozaki K, Karpanen T, et al: Suppression of tumor lymphangiogenesis and lymph node metastasis by blocking vascular endothelial growth factor receptor 3 signaling. J Natl Cancer Inst 94::819,2002-825, Crossref, Medline, Google Scholar |
| 57. | Shimizu K, Kubo H, Yamaguchi K, et al: Suppression of VEGFR-3 signaling inhibits lymph node metastasis in gastric cancer. Cancer Sci 95::328,2004-333, Crossref, Medline, Google Scholar |
| 58. | Jain RK, Fenton BT: Intratumoral lymphatic vessels: A case of mistaken identity or malfunction? J Natl Cancer Inst 94::417,2002-421, Crossref, Medline, Google Scholar |
| 59. | Padera TP, Kadambi A, di Tomaso E, et al: Lymphatic metastasis in the absence of functional intratumor lymphatics. Science 296::1883,2002-1886, Crossref, Medline, Google Scholar |
