^{1}

^{2}

^{*}

Background: Regional lymph node invasion and metastasis may happen early during the progres-sion of gastric cancer. The lymphadenectomy is still the key method to treat lymph nodemetastasis. In the recent years, scientists have found some growth factors and receptors that can promote angiogenesis which also play an important role in adjusting the formation of the new lymph vessel, and lymphangiogenesis may play a key role in the process of lymph node metastasis. Objectives: This study aims to explore the correlation between the expression of vascular endothelial growth factor-C (VEGF-C), vascular endothelial growth factor receptor 3 (VEGFR-3) and lymph node me-tastasis (LNM), and its impact on prognosis of patients with gastric cancer. Methods: The samples were collected from gastric cancer database of Sichuan Provincial People’s Hospital from 2005 to 2007, which were registered and followed up. The samples were divided into two groups according to situation whether there is lymph node metastasis, which is lymph node metastasis N(+) and without lymph node metastasis N(﹣). The expression of VEGF-C, VEGFR-3 and CD34 were measured by immuno histochemistry staining with monoclonal antibody (anti-VEGF-C, anti-VEGFR-3, and anti-CD34). Kaplan-meier, logistic and Cox regression was performed to explore their impact on the prognosis of patients with gastric cancer. Results: In total 186 cases were collected, 96 cases in N(+) group and 90 cases in N(﹣) group. The percentage of VEGF-C expression is 54.83% (102/186) in all groups, 73.9% (71/96) in N(+) group, and 34.44% (31/90) in N(﹣) group ( p = 0.001). The percentage of VEGFR-3 expression is 33.33% (62/186) in all groups, 44.78% (43/96) in N(+) group, and 21.11% (19/90) in N(﹣) group ( p = 0.001). There are no statistical differences in microvessel density (MVD) between N(﹣) and N(+) group. The average lymphatic vessel density (LVD) was significant different between N(+) and N(﹣) group (26.23 ± 8.2 and 18.46 ± 7.4, t = ﹣2.427, p = 0.016). The five-year overall survival rate of N(+) group is 31% and the N(﹣) group is 66%; there are statistical differences between the two groups (Log rank = 27.15, p = 0.001). The five-year overall survival rates of VEGF-C positive group and VEGF-C negative group are 36% and 59%, with the statistical differences (Log rank = 27.15, p = 0.001). And the five-year overall survival rates of VEGFR-3 positive group and VEGFR-3 negative group are 31% and 43%, also with the statistical differences (Log rank = 5.241, p = 0.041). Conclusions: The expressions of VEGF-C, VEGFR-3 in cell plasma of gastric cancer tissue not only correlate with lymphatic vessel density and lymph node metastasis (LNM), but also are important factors which impact prognosis of gastric cancer patients.

Gastric cancer is one of the most common malignant tumors in the developing countries, whose incidence ranks the second place, only after lung cancer in the world [

In the process of gastric cancer, the most common metastasis mode is lymph node metastasis. Epidemiological statistical analysis found that even in early gastric cancer patients, the lymph node metastasis rate can reach about 10% - 15%. The scope and number of the lymph node metastasis decide the scope of lymphadenectomy in gastric cancer operations and directly affect the postoperative prognosis of the gastric cancer patients. In the recent years, with the non-stop development of research related to molecular biology, scientists have found some growth factors and receptors that can promote angiogenesis which also play an important role in adjusting the formation of the new lymph vessel and lymphangiogenesis may play a key role in the process of lymph node metastasis after the malignant progression [

Current with the analysis of the expression of VEGF-C in the gastric cancer tissue and VEGFR-3 of the tumor edge and the correlation with lymph node metastasis, this study will explore its relationship with the formation of the new lymph vessel and tumor lymph node metastasis in order to analyze its impact on the prognosis of gastric cancer and preliminarily discuss the application value of the new lymph vessel targeted biological therapy.

・ Monoclonal antibody CD34: purchased from NEOMAKER, America;

・ Monoclonal antibody VEGF-C: purchased from NEOMAKER, America;

・ Monoclonal antibody VEGFR-3: purchased from NEOMAKER, America;

・ Coomassie brilliant blue solution: purchased from LAB VISION, America

・ 10% neutral formalin: formalin liquid, 10 ml & 0.01 mol/L PBS, 90 ml;

・ 0.01 mol/L phosphate buffer saline (PBS) solution;

・ Antigen retrieval solution: A solution (0.1 M/L citrate); B solution (0.1 M/L sodium citrate);

・ Working solution: A solution of 9 ml with B solution of 41 ml in addition to distilled water of 450 ml;

・ Diaminobenzidine (DAB) immunohistochemistry kit: purchased from LAB VISION, America;

・ Streptavidin-Peroxidase (SP) immunohistochemistry reagent: purchased from Maifa Biological Technology Company, China.

・ Paraffin embedding machine: purchased from Sheldon, America;

・ Letz1512 paraffin slicing machine: purchased from Leica, Germany;

・ OLYMPUS upright microscope: purchased from Olympus, Japan.

The test specimen is the gastric cancer tissue specimens of the gastrointestinal surgical resection from the Sichuan Provincial People’s Hospital from April 2005 to December 2007. According to the database record of the gastric tissue and follow-up visits, there are a total of 186 cases of paraffin-embedded specimens. Based on the situation whether there is lymph node metastasis, the specimens are divided into two groups: lymph node metastasis N(+) and without lymph node metastasis N(−).

The gastric cancer tissues are embedded in paraffin blocks, sliced into sections of 3 μm thickness, attached to clean poly-lysine-coated films and then put into the clip box at room temperature for future use.

Regular dewaxed to water: Slice the tissue embedded in paraffin and put them into an oven of 65˚C and bake overnight. Place them into xylene to dewax on the next day, 10 minutes in xylene I and xylene II respectively, following the step of gradient alcohol hydration, 2 minutes every time in ethanol for twice and then 2 minutes in 95% ethanol, 80% ethanol, 70% ethanol and water respectively.

Put the slices into the hematoxylin staining solution for 5 to 10 minutes.

Gently rinse the slices with tap water for 5 to 10 minutes.

Acidify the slices in the 2% hydrochloric acid alcohol for 10 seconds and then gently rinse them with tap water for about 10 minutes.

Put the slices in the 1/500 ammonia until the slices turn blue for 20 seconds and then immediately gently rinse them with tap water for 4 - 5 minutes (with distilled water once).

Dye the slices in the 0.5% - 1% eosin staining solution for 3 - 5 minutes and gently rinse them with tap water once.

And then it is gradient alcohol dehydration with 80% alcohol, 95% alcohol and 100% alcohol for 2 minutes each, followed by put them in xylene I and II for 10 minutes each.

Use neutral gum to fix the slices.

Regular dewaxed to water: Slice the tissue embedded in paraffin and put them into an oven of 65˚C and bake overnight. Place them into xylene to dewax on the next day, 10 minutes in xylene I and xylene II respectively, following the step of gradient alcohol hydration, 2 minutes every time in ethanol for twice and then 2 minutes in 95% ethanol, 80% ethanol, 70% ethanol and water respectively.

Configurate fresh antigen retrieval solution for paraffin slices to perform antigen retrieval.

Put the slices into PBS solution and then gently wash for five minutes on the vibrating shaker which shall be repeated three times.

Configurate peroxidase blocking solution, and endogenously block the paraffin slices for 10 minutes.

Put the slices into PBS solution and then gently wash for five minutes on the vibrating shaker which shall be repeated three times.

Configurate fresh 10% non-immune system animal blood serum blocking solution and block the nonspecific antigen of the paraffin slices for 5 minutes.

Put the slices into PBS solution and then gently wash for five minutes on the vibrating shaker which shall be repeated three times.

According to the primary antibodies manual, respectively anti-dilute and configurate fresh primary antibodies working solution.

Add the primary antibodies and put the slices in a 4˚C refrigerator overnight. The next day, get the slices out at room temperature for an hour.

Configurate biotinylated secondary antibodies working solution 50 ml and drop it on the slices for 10 minutes’ incubation.

Add 50 microliters of streptavidin peroxidase and incubate for 10 minutes.

Configurate 100 microliters fresh DAB TMB substrate solution and render at room temperature for 3 - 5 minutes (Perform a real-time observe the rendering process under the microscope).

Finally, redye the slices with hematoxylin solution.

And then it is gradient alcohol dehydration with 80% alcohol, 95% alcohol and 100% alcohol for 2 minutes each, followed by put them in xylene I and II for 10 minutes each.

Use neutral gum to fix the slices.

Replace the primary antibodies working solution with PBS as the negative blank contrast and replace the secondary antibodies working solution with normal serum as the blank contrast. An anti-dilution ratio of the primary antibodies working fluid is as follows in order: VEGF-CIg (McAb) 1:100, VEGFR-3 (McAb) 1:100, CD34 (McAb) 1:100.

Under the microscope, the positive expression of VEGF-C shows the rendered color of cytoplasm is brown; the positive expression of VEGFR-3 shows the rendered color of membrane is brown; the positive expression of CD34 shows the rendered color of membrane and cytoplasm is brown.

With a 100× microscope, calculate the cell numbers among 100 cancer cells which show the positive expression of VEGF-C or VEGFR-3 in five observations and calculate the ratio of positive expression cancer cells. Above 20% is the positive expression; 20% - 40% is +; 41% - 60% is ++; 61% - 80% is +++; and above 81% is ++++.

Assessment of the staining result of CD34 and microvessel density (MVD) assessment: the staining result demonstrates a significant difference from normal epithelial cancer cells and other tissue components, among which the stained brown cells or cell clusters are assessed as a blood vessel or a branch structure not connected can be calculated as a blood vessel. The immunohistochemistry slices are first put under a low-magnification optical microscope to determine the three highest vascular density areas. And then use a high-magnification optical microscope to choose the four highest vascular density areas to count the blood vessels (The average size of every field of observation is 0.075 mm^{2}). Take the average value of the average blood vessel number as the specimen and the calculation method of MVD is the same.

At least two or more pathologists observe the immunohistochemistry slices in a double-blind reading manner and record the number and distribution of cells that express positively and pathological type and immunohistochemical features, etc.

All the experimental data are processed and analyzed by the SPSS17.0 statistical software. The experimental results are expressed as mean ± standard deviation (X ± SD), count data using the X^{2} test, measurement data using t test. Between the positive and negative expression groups of VEGF and VEGFR-3, all the factors use X^{2} test and t test. The relationship between lymph node metastasis status and various factors is analyzed with the use of bivariate logistics regression analysis; calculate the overall survival rate using Kaplan-Meier; compare the survival rate curves using Log rank and Breslow; the prognostic factors are analyzed using Cox regression, where p < 0.05 is considered to have the statistical meaning.

Based on the situation of whether there is lymph node metastasis, divide the 186 cases into two groups, the positive lymph node metastasis N(+) and the negative lymph node metastasis N(−). There are no statistical differences between the two groups in terms of gender, age, tumor location, gross type, tumor size and histopathology, etc.; there are statistical differences in terms of the depth of invasion. Among the 186 cases, 90 are N(−) and 96 are N(+); the positive expression of VEGF-C is 54.83% (102/186), among which, 34.44% (31/90) is N(−) and 73.9% (71/96) is N(+), where p = 0.001. In the total specimen, the positive expression of VEGFR-3 is 33.33% (62/186), 21.11% (19/90) as N(−) and 44.78% (43/96) as N(+), where p = 0.001, shown in

The positive expression rates of VEGF-C and VEGFR-3 both grow with the increasing clinical stages. The positive expression rate differences of VEGF-C in different clinical stages have statistical significance (p < 0.05); the positive expression rate differences of VEGFR-3 in different clinical stages do not have statistical significance, shown in

The expression of VEGF-C in cytoplasm of gastric cancer is shown as

To evaluate microvessel density (MVD) with CD34, there are no statistical differences between N(−) and N(+). To evaluate lymphatic vessel density (LVD) with the staining situation of microlymphatic vessel, LVD of N(−) is 18.46 ± 7.4; LVD of N(+) is 26.23 ± 8.2, with a significant difference (t = −2.427, p = 0.016). Divide the two groups of cases based on whether there is the expression of VEGF-C and there are no statistical differences of the average MVD between the two groups and there are statistical differences of the average LVD between the two groups, respectively 18.12 ± 10.01 and 25.89 ± 13.28 (t = −2.449, p = 0.015), shown in

The expression of CD34 in vascular endothelial cell membrane and cytoplasm is shown as

Cases | |||
---|---|---|---|

Clinicopathologic factors | Lymph node metastasis (n = 96) | Non-lymph node metastasis (n = 90) | p value |

Gender | 0.355 | ||

Male | 74 | 65 | |

Female | 22 | 25 | |

Age | 0.388 | ||

≤60 | 52 | 56 | |

>60 | 44 | 34 | |

Tumor location | 0.412 | ||

Upper stomach 1/3 | 25 | 17 | |

Middle stomach 1/3 | 16 | 17 | |

Lower stomach 1/3 | 55 | 56 | |

Gross type | 0.233 | ||

Borrmann I | 6 | 8 | |

Borrmann II | 26 | 32 | |

Borrmann III | 57 | 41 | |

Borrmann IV | 7 | 9 | |

Tumor size | 0.288 | ||

≤4 cm | 59 | 63 | |

>4 cm | 37 | 27 | |

Tumor depth | 0.001 | ||

pT1 | 7 | 34 | |

pT2 | 21 | 27 | |

pT3 - 4 | 68 | 29 | |

Differentiation degree | 0.155 | ||

High | 4 | 13 | |

Medium | 30 | 26 | |

Low | 62 | 51 | |

Clinical stage | 0.0001 | ||

I | 7 | 45 | |

II | 17 | 25 | |

III | 52 | 17 | |

IV | 20 | 3 | |

Expression of VEGF-C | 0.001 | ||

Positive | 71 | 31 | |

Negative | 25 | 59 | |

Expression of VEGFR-3 | 0.001 | ||

Positive | 43 | 19 | |

Negative | 53 | 71 |

Clinicopathologic factors | Cases | Expression of VEGF-C | p value | Expression of VEGFR-3 | p value | ||
---|---|---|---|---|---|---|---|

Positive | Negative | Positive | Negative | ||||

Gender | 0.231 | 0.228 | |||||

Male | 139 | 80 | 59 | 50 | 89 | ||

Female | 47 | 22 | 25 | 12 | 35 | ||

Age | 0.849 | 0.329 | |||||

≤60 | 108 | 60 | 48 | 33 | 75 | ||

>60 | 78 | 43 | 35 | 30 | 48 | ||

Tumor location | 0.061 | 0.628 | |||||

Upper stomach 1/3 | 42 | 32 | 10 | 17 | 25 | ||

Middle stomach 1/3 | 33 | 12 | 21 | 11 | 22 | ||

Lower stomach 1/3 | 111 | 59 | 52 | 35 | 76 | ||

Gross type | 0.012 | 0.777 | |||||

I | 14 | 10 | 4 | 6 | 8 | ||

II | 59 | 27 | 32 | 18 | 41 | ||

III | 96 | 61 | 35 | 36 | 60 | ||

IV | 17 | 5 | 12 | 3 | 14 | ||

Tumor depth | 0.028 | 0.122 | |||||

pT1 | 40 | 17 | 23 | 10 | 30 | ||

pT2 | 49 | 24 | 25 | 13 | 36 | ||

pT3 - 4 | 97 | 62 | 35 | 39 | 58 | ||

Differentiation degree | 0.614 | 0.244 | |||||

High | 16 | 7 | 9 | 4 | 12 | ||

Medium | 57 | 30 | 27 | 16 | 41 | ||

Low | 113 | 65 | 48 | 43 | 70 | ||

Clinical stage | 0.002 | 0.072 | |||||

I | 52 | 24 | 28 | 14 | 38 | ||

II | 41 | 15 | 26 | 10 | 31 | ||

III | 70 | 47 | 23 | 29 | 41 | ||

IV | 23 | 17 | 6 | 11 | 12 | ||

Lymph node metastasis | 0.0001 | 0.002 | |||||

Positive | 96 | 71 | 25 | 43 | 53 | ||

Negative | 90 | 31 | 59 | 19 | 71 |

Note: X^{2} test.

In the experiment, the five-year overall survival rate of the positive group N(+) of lymph node patients is 31% and the negative group is 66% so that there are statistical differences between the two groups (Log rank = 27.15, p = 0.001). To divide the cases based on the situation whether there is the expression of VEGF-C, the five-year overall survival rates of the two groups are 36% (the expression of VEGF-C is positive) and 59% (the expression of VEGF-C is negative) with the statistical differences (Log rank = 27.15, p = 0.001); to divide the cases based on the situation whether there is the expression of VEGFR-3, the five-year overall survival rates of the two groups are 31% (the expression of VEGFR-3 is positive) and 43% (the expression of VEGFR-3 is negative), with the statistical differences (Log rank = 5.241, p = 0.041). In the further subgroup analysis, in the group of positive expression of VEGF-C, the five-year overall survival rate of non-expression of VEGFR-3 is 43% and the five-year overall survival rate of expression of VEGFR-3 is 31%, with the statistical differences (Log rank = 2.273, p = 0.032); and in the group of negative expression of VEGF-C, the five-year overall survival rate of the negative expression of VEGFR-3 is 66% and the five-year overall survival rate of the positive expression of VEGFR-3 is 44%, with the statistical differences (Log rank = 0.001, p = 0.697), see

In the binary logistic regression analysis, the lymph node metastasis is the dependent variable, among which, the negative lymph node metastasis is set to be 0 and the lymph node metastasis is set to be 1; the independent variables include gender of patients, age, Borrmann type, tumor location, invasion depth, differentiation degree, TNM stage, invasion to adjacent organs, and the expression of CD34, VEGF-C and VEGFR-3. Backward: use the LR method to select the covariates and the inspection deletion covariate standard is α = 0.10 (default value). After Step 13 selection process, delete gender, age, tumor location, Borrmann type, invaded adjacent organs and CD34 (MVD) of the covariates and ultimately retain covariates in the model as the analysis results, which are VEGF-C, VEGFR-3, differentiation degree, invasion depth and TNM stage, see in

In the Cox regression analysis of 186 gastric cancer patients, the dependent variable is the survival time of the follow-up patients; the independent variables are consisted of the patients’ gender, age, tumor location, Bor- rmann type, differentiation degree, invasion depth, lymph node metastasis, lymph node metastasis station, liver metastasis, lymph node dissection scope, combined organ resection, operative time, blood transfusion, tumor

Indector | Five-year survival rate | Logrank Chi-square | p | Breslow Chi-square | p |
---|---|---|---|---|---|

Lymph node | 27.15 | 0.0001 | 29.53 | 0.0001 | |

LN(−) | 66% | ||||

LN(+) | 31% | ||||

VEGF-C | 8.638 | 0.003 | 8.183 | 0.004 | |

VEGF-C(−) | 59% | ||||

VEGF-C(+) | 36% | ||||

VEGFR-3 | 5.241 | 0.041 | 5.263 | 0.0423 | |

VEGFR-3(−) | 50% | ||||

VEGFR-3(+) | 42% | ||||

VEGF-C(+) | 2.273 | 0.032 | 2.281 | 0.031 | |

VEGFR-3(−) | 43% | ||||

VEGFR-3(+) | 31% | ||||

VEGF-C(−) | 0.001 | 0.697 | 0.219 | 0.639 | |

VEGFR-3(−) | 66% | ||||

VEGFR-3(+) | 44% |

Step 13 | B | S.E | Wald | df | Sig | Exp(B) | 95.0% C.I. for Exp(B) | |
---|---|---|---|---|---|---|---|---|

Lower | Upper | |||||||

VEGF-C | 1.544 | 0.435 | 12.613 | 1 | 0.0001 | 4.684 | 1.988 | 10.982 |

VEGFR-3 | 0.342 | 0.105 | 10.534 | 1 | 0.001 | 1.408 | 1.145 | 1.730 |

Invasion | 0346 | 0.356 | 9.436 | 1 | 0.033 | 1.708 | 1.352 | 1.402 |

Grade | 0.249 | 0.302 | 10.817 | 1 | 0.093 | 1.002 | 1.154 | 1.799 |

TNM stage | 0.942 | 0.172 | 31.044 | 1 | 0.000 | 2.566 | 1.832 | 3.595 |

Constand | −4.362 | 0.705 | 38.266 | 1 | 0.000 | 0.013 |

size, TNM stage, expression of VEGF-C and VEGFR-3 and microvessel density. By using the forward stepwise covariates, the standard of covariates is α = 0.05 (default value) and it has to take 13 steps, including the expression of VEGFR-3 and VEGF-C, lymph node metastasis, liver metastasis, invasion depth, lymph node dissection scope, intraoperative blood transfusion, and the age of patients ultimately retained in the model covariates. The results show that the OR values of invasion depth of tumors, lymph node metastasis, liver metastasis, intraoperative blood transfusion, and the expression of VEGF-C and VEGFR-3 are all greater than 1 so that they are the dangerous factors to reduce the patients’ survival time, but the OR values of the lymph node dissection scope and the patients’ ages are both smaller than 1 so that they are the protective factors to increase the patients’ survival time. The final risk function is

see

Step 10 | B | SE | Wald | df | Sig | Exp(B) | 95.0% CI for Exp(B) | |
---|---|---|---|---|---|---|---|---|

Lower | Upper | |||||||

Lymph node metastasis | 0.726 | 0.273 | 7.095 | 1 | 0.008 | 2.066 | 1.211 | 3.525 |

Invasion | 0.691 | 0.162 | 18.168 | 1 | 0.000 | 1.996 | 1.452 | 2.742 |

VEGFR-3 | 0.077 | 0.036 | 4.680 | 1 | 0.031 | 1.080 | 1.07 | 1.158 |

VEGF-C | 0.469 | 0.177 | 7.009 | 1 | 0.008 | 1.599 | 1.130 | 2.263 |

Hepatic metastasis | 0.704 | 0.363 | 3.749 | 1 | 0.053 | 2.021 | 0.991 | 4.122 |

Lymph node dissection | −0.403 | 0.150 | 7.195 | 1 | 0.007 | 0.668 | 0.497 | 0.897 |

Age | −0.345 | 0.113 | 9.401 | 1 | 0.002 | 0.708 | 0.568 | 0.883 |

Volume of blood transfusion | 0.01 | 0.000 | 25.534 | 1 | 0.000 | 1.001 | 1.001 | 1.002 |

The death rate of gastric cancer ranks first among a variety of tumors and thus it is imperative for the prevention of gastric cancer. Clinical studies have shown that after the perform of radical gastrectomy, the five-year survival rate of patients with early gastric cancer is 90%, while the rate of patients with advanced is only 40%, among which some patients die of local recurrence and metastasis. This has also shown that the metastasis which cannot be determined by the current morphological diagnostic techniques during the process of treatment, which is micrometastasis. A variety of clinicopathological studies have indicated that lymph node metastasis is one of the major ways for the solid tumors to disseminate; in the early stages of gastric cancer, the dissemination of the regional lymph node for the tumor cells already exist. In the recent years, the continuous findings by researchers and studies of the specific marker of the lymphatic endothelial cell are helpful for the basic study of the gastric cancer lymphangiogenesis and deep exploration of the molecular mechanisms of the lymph node metastasis, which not only can provide an important basis to determine the prognosis for the gastric cancer patients, but also can provide a more solid theoretical basis for the future targeted therapies.

Regarding tumor metastasis, people always focus on tumor angiogenesis and new vessel research without the emphasis on the research of tumor lymphangiogenesis, the main reason of which is that there is lack of specificity of the identification of the markers of lymphangiogenesis. Regarding the phenomenon that the lymph vessels surrounding the tumor often expand and penetrate the tumor, it has been unclear whether the expanded lymph vessels are the lymphangiogenesis after the tumor or before the tumor. Some scholars think the increased mesenchyme pressure within the tumor differ the lymph vessels’ structure and functions within the tumor from the surrounding normal lymph vessels so that there are lack of functional lymph vessels within the tumor. Recent studies have demonstrated that with the expression of the lymphangiogenesis factors VEGF-C and VEGF-D, etc., especially by VEGF-C, VEGF-D/VEGFR-3 pathway, the cancer cells can induce lymphangiogenesis and further promote cancer lymphatic metastatic cells. These findings make lymphangiogenesis become the research focus of studying lymphatic metastasis and there may be in the treatment of lymphatic metastasis of tumor targets [

In 1983, Senger et al. [

In the recent years, there have been many studies about the relationship of the formation of new blood vessels and lymphatic vessels with VEGF pathway and the functions in the process of tumor growth, which have proved that they are closely associated with the growth and metastasis of tumors. In 1996, in the use of optical chromatography, Joukov et al. firstly isolated and cloned VEGF-C, the new member of VEGF/PDGF family from human prostate cancer cell line PC-3 cDNA library [^{4}, 2.9 × 10^{4} and connected by a disulfide bond. When it is secreted outside the cell, it can be immediately hydrolyzed into a small subunit with a molecular weight of 2.1 × 10^{4} and not linked by a disulfide bond by plasminogen and relevant proteolytic enzymes [

Lymphangiogenesis is the process of further development and reproduction of the original lymphatic vessels to new lymphatic vessels. The process exists in the conditions of embryonic growth, inflammation, tissue repair and parasitic infections under, etc. [

There are many hypotheses about the original of lymphatic vessels, and right now there are two major points of view. One hypothesis is that the lymphatic vessels are differentiated from the stroma precursor cells―em- bryonic cells of lymphatic vessels within mesenchyme; the other hypothesis is that the lymphatic network comes from the vesicle of the embryonic vein endothelial cells in the way of regional budding. With the reveal of the specific molecular markers of the lymphatic vessels and the corresponding growth elements, the exploration of the lymphangiogenesis molecular mechanism has entered into a new phase. The over expression of VEGF-C can promote the continuous lymphangiogenesis and lead to lymph node metastasis. VEGFR-3 (Flt-4) is mainly expressed in adults’ lymphatic endothelium cell [

Lymphangiogenesis induced by the tumor promotes the occurrence of tumor cells’ lymph node metastasis, wherein the endogenous VEGF-C is an important regulator of lymphangiogenesis which plays an important role. Based on the current series of experiments in animal models, VEGF-C plays an important role onlymphangiogenesis and tumor metastasis. Skobeand Yanai et al. [

Currently, it is not very clear about the signal of VEGF-C/VEGFR-3 in tumor lymphangiogenesis and in the regulation mechanism of the precise factors in lymph node metastasis. However, it is clear that the lymphatic endothelial cells indeed participate in the generation of tumor lymphangiogenesis and lymph node metastasis. Its molecular mechanisms may be: 1) The structural features of lymphangiogenesis make it easier for the tumor cells to invade into the lymphatic system. Under an electron microscopy, the tumor lymphatic wall is irregular, endothelial cells swollen, organelles abnormal morphological, the connections of endothelial cells not tight and in a state of expansion open. 2) Some scholars believe the expression of VEGF-C in tumor cells and some chemotactic cytokines of biochemical reactions both participate in tumor cell metastasis; maybe the change of the activity of some adhesion molecules on lymphatic vessels can make thechemotactic ability of tumor cells and invasion power of lymphatic vessels stronger [

The expression of VEGF-C and its specific receptor VEGFR-3 in prostate cancer cells plays an important role on the overall development of prostate cancer. In corresponding studies of human prostate cancer, Tsurusaki et al. used in situ hybridization and found out VEGF-C mainly promoted lymphangiogenesis network in tumor tissues by its specific combination with its receptor VEGFR-3 and accordingly brought in lymph node metastasis [

Among the 186 gastric cancer tissue specimens, the positive and negative lymph node metastasis cases are 96 and 90. The positive expression rate of VEGF-C is 54.83% (102/186), where N(−) group is 34.44% (31/90) and N(+) is 73.9% (71/96) (p = 0.001). The immunohistochemical staining of lymphatic vessels on the edge of the tumor show the positive expression of VEGFR-3 is about 33.33% (62/186), where N(−) is 21.11% (19/90) and N(+) is 44.78% (43/96) (p = 0.001) with statistical differences. The lymph node metastasis of positive expression of VEGF-C is about 68.93% (71/103) and the negative expression lymph node metastasis rate is about 28.91% (24/83), p = 0.001; the positive expression lymph node metastasis rate is about 66.66% (42/63) and positive expression rate is about 43.09% (53/123), p = 0.002 with statistical differences. The result shows the expression of VEGF-C and VEGFR-3 is closely related to tumor lymph node metastasis. Interestingly, the expression rates of VEGF-C and VEGFR-3 are increased with the growth of clinical stages and the positive expression rate of VEGF-C has statistical differences among various clinical stages (p < 0.05), but the positive expression rates of VEGFR-3 among various clinical stages do not have statistical differences. Evaluating the microvessel density (MVD) through CD staining, there are no statistical differences between the positive and negative groups of lymph node. Through the VRGFR-3 staining of microlymphatics, assessing lymphatic vessel density (LVD), there are significant statistical differences between the positive and negative groups of lymph node; average LVD shows significant differences. This shows that gastric cancer lymph node metastasis is obviously related to the expression level of VEGF-C. In terms of lymph node metastasis, LVD is a more accurate index. As known to all, gastric cancer lymph node metastasis is one of the major elements of affecting prognosis. In this study, patients of the lymph node positive group only have 31% of five-year overall survival rate; the five-year overall survival rate for the negative group patients reaches 66% with significant statistical differences. The five-year overall survival rate of the negative expression of VEGF-C patients is about 59%; the five-year overall survival rate of the positive expression patients is about 36% with statistical differences. The five-year overall survival rate of the negative expression of VEGFR-3 patients is about 43%; the five-year overall survival rate of the positive expression patients is only about 31% also with statistical differences. A further stratified analysis has found out that among the positive expression of VEGF-C and VEGFR-3 patients the five-year overall survival rate is only 31% with statistical differences, indicating a poor prognosis. The logistic regression analysis has found out that the OR vales of VEGF-C and VEGFR-3 are both greater than 1 to show that they are dangerous elements for gastric cancer lymph node metastasis. The Cox regression analysis has found out the OR values of tumor invasion depth, lymph node metastasis, liver metastasis, intraoperative blood transfusion and the VEGF-C and VEGFR-3 expression are greater than 1 to show they are dangerous elements to reduce patients’ survival time. The OR values of lymph node dissection and patients ages are smaller than 1 so that they are protective elements to increase patients’ survival time. Based on the above experiments’ results, it is believed that the expressions of VEGF-C and its specific receptor VEGFR-3 in gastric cancer are associated with lymph node metastasis. There is a poor prognosis for the patients when the expressions of VEGF-C and VEGFR-3 are positive. The expression of VEGF-C is one of the independent indexes to predict lymph node metastasis, which can be used as an independent element to affect the prognosis of the gastric cancer patients.

In summary, our data demonstrated that the expressions of VEGF-C, VEGFR-3 in cell plasma of gastric cancer tissue not only correlate with lymphatic vessel density and lymph node metastasis (LNM), but also are important factors which impacts prognosis of gastric cancer patients.

We thank the participants and their families for their kind cooperation, generosity and patience.