- Email: [email protected]
Clinical significance of urinary vascular endothelial growth factor and microvessel density in patients with renal cell carcinoma
ADULT UROLOGY
CLINICAL SIGNIFICANCE OF URINARY VASCULAR ENDOTHELIAL GROWTH FACTOR AND MICROVESSEL DENSITY IN PATIENTS WITH RENAL CELL CARCINOMA SUNG-GOO CHANG, SUNG-HYUN JEON, SUN-JU LEE, JOONG-MYUNG CHOI, YOUN-WHA KIM
AND
ABSTRACT Objectives. To investigate the urinary vascular endothelial growth factor (VEGF) levels from patients with renal cell carcinoma (RCC). Neovascularization, an essential event for the growth of solid tumors, is regulated by a number of angiogenic factors. VEGF is thought to exert potent angiogenic activity. Methods. Urine samples were obtained before radical nephrectomy from 27 patients with RCC and 10 control subjects with no evidence of cancer or inflammatory disease. VEGF was measured by enzyme-linked immunosorbent assay in the urine and corrected according to the 24-hour urine concentration of creatinine. The microvessel density was measured by immunohistochemical staining with CD31 monoclonal antibody. Nuclear morphometry was performed by photomicroscopy. Results. The corrected urinary VEGF levels in patients with RCC were much higher than those in the normal control group (P ⫽ 0.039) and were more elevated in patients with higher stages of RCC (Stages III and IV versus Stages I and II; P ⫽ 0.024). A tendency was also noted for the VEGF levels to be higher according to cell grade. However, no statistical correlation was found between the corrected urinary VEGF and age, sex, tumor size, cell type, microvessel density, platelet count, or hemoglobin. The nuclear area was higher with more advanced-stage tumors (P ⫽ 0.043) and tended to increase according to the tumor cell grade. Conclusions. The results of this study indicate that urinary VEGF levels are increased in patients with RCC. However, they may not reflect the underlying angiogenic activity, and it may be that other angiogenic factors play a more prominent role. UROLOGY 58: 904–908, 2001. © 2001, Elsevier Science Inc.
N
eovascularization is essential to the growth and metastasis of solid tumors.1 Previous studies have shown that angiogenic factors are involved in the neovascularization of renal cell carcinoma (RCC),2 brain tumors,3 and colon cancer.4 The microvessel density in the tumor tissue correlated with the malignant potential of some solid tumors.5 RCCs are solid tumors characterized by abundant neovascularization and arteriovenous fistula formation.6 The tumor vascularity is related to the clinical outcome, because metastases are more likely in patients with highly vascularized RCC.7 From the Departments of Urology, Preventive Medicine, and Pathology, Kyung Hee University School of Medicine, Seoul, Korea Reprint requests: Sung-Goo Chang, M.D., Ph.D., Department of Urology, Kyung Hee University Medical Center, No. 1 Hoegidong Dongdaemun Ku, Seoul 130-702, Korea Submitted: May 18, 2001, accepted (with revisions): July 24, 2001
904
© 2001, ELSEVIER SCIENCE INC. ALL RIGHTS RESERVED
It has been proposed that tumor-induced angiogenesis is stimulated in a paracrine fashion by growth factors produced by tumor cells that bind to vascular endothelial cells, resulting in cellular proliferation.8 In contrast to other angiogenic growth factors, vascular endothelial growth factor (VEGF) is a secreted peptide with activity restricted to vascular endothelial cells.9 Increased serum levels of VEGF from patients with RCC have been reported,10 and these levels were also increased in RCC tissue.11 Crew et al.12 quantified VEGF in the urine of patients with bladder cancer. This study was undertaken to determine, through enzyme-linked immunosorbent assay and immunohistochemistry, the quantitation of urinary VEGF in patients with RCC and a possible relationship between the urinary VEGF levels and disease characteristics such as microvessel density, nuclear area, tumor grade, and tumor size. 0090-4295/01/$20.00 PII S0090-4295(01)01375-9
TABLE I. Patient characteristics Patients (n) Men
16
Women
11
Control
9
Corrected Urinary VEGF*
P Value
0.195 ⫾ 0.381 (0.014–1.032) 0.070 ⫾ 0.114 (0.009–1.204) 0.260
RCC
27
0.024 ⫾ 0.026 (0.000–0.040) 0.147 ⫾ 0.305 (0.009–1.204)
TABLE II. Levels of corrected urinary VEGF according to pathologic characteristics of the RCC Patients (n)
Corrected Urinary VEGF
Stage Low (T1, T2)
18
High (T3, T4)
9
0.050 ⫾ 0.045 (0.009–0.158) 0.340 ⫾ 0.480 (0.007–1.204)
P Value
0.024 0.039
KEY: VEGF ⫽ vascular endothelial growth factor; RCC ⫽ renal cell carcinoma. Data presented as the mean ⫾ SD, with the range in parentheses, unless noted otherwise. * Corrected urinary VEGF ⫽ level of urinary VEGF/24-hour urine creatinine.
Pathologic grade 1
2
2
17
3
2
0.041 ⫾ 0.048 (0.007–0.075) 0.069 ⫾ 0.098 (0.009–0.403) 0.623 ⫾ 0.821 (0.007–1.204) 0.009*
MATERIAL AND METHODS PATIENTS The patients included 16 men and 11 women (mean age 52.6 years). We analyzed 37 urine samples from 27 patients with RCC and 10 patients with ureteral stones as controls. The patients were treated with radical nephrectomy (21 cases), had a diagnosis only (4 cases), and had an angioinfarction (2 cases). The RCC stage was classified by TNM classification and the grade according to Fuhrman.
Size (cm) ⱖ7 ⬍7
13 14
0.560 Nuclear area (m2) 0–43.2 ⬎43.2
11 10
ENZYME-LINKED IMMUNOSORBENT ASSAY Urine VEGF was measured by the Quantikine (R&D System, Minneapolis, Minn) enzyme-linked immunosorbent assay. The measurable range of this method was from 15.6 to 1000 pg/mL. To control for differences in the urine concentration, urinary VEGF was expressed relative to the 24-hour urinary creatinine content.
IMMUNOHISTOCHEMISTRY Immunohistochemical staining was performed using a streptavidin-biotin immunoperoxidase method according to the supplier’s protocol (Dako, LSAB Kit, Carpenteria, Calif). In brief, paraffin-embedded sections were deparaffinized in xylene and rehydrated with graded ethanol. After quenching the endogenous peroxidase activity in 0.3% hydrogen peroxide for 30 minutes and blocking reagents for 30 minutes, primary monoclonal mouse anti-human endothelial cell, CD31 (Dako), was applied to the sections at a dilution of 1:20 and incubated in a moist chamber for 2 hours at room temperature. After washing out the excess complex, the localization of antibodies was visualized by incubating the section for 10 minutes in 3,3⬘-diaminobenzidine tetrahydrochloride (Research Genetics, Huntville, Ala). The positive controls for human erythrocyte glucose transporter (Glut-1) were the red blood cells present in each section. The negative control sections were made of horse serum without the primary Glut-1 UROLOGY 58 (6), 2001
0.038 ⫾ 0.034 (0.007–0.110) 0.200 ⫾ 0.369 (0.009–1.204) 0.205
URINE COLLECTION A midstream urine specimen (10 mL) was collected from patients before treatment, and the creatinine concentration in the 24-hour collected urine specimen was measured by the Jaffe reaction method. The collected urine for measurement of VEGF was centrifuged at 3000g for 3 minutes. The supernatant was separated and stored at ⫺70°C until testing.
0.058 ⫾ 0.048 (0.009–1.204) 0.042 ⫾ 0.043 (0.014–0.770)
KEY: Abbreviations as in Table I. Data presented as the mean ⫾ SD, with the range in parentheses, unless noted otherwise. * Tested by analysis of variance.
antibody. One pathologist who was unaware of the pathologic findings independently interpreted the stained sections. Vessels with a clearly defined lumen or well-defined linear vessel shape that stained brown, but not single endothelial cells, were counted to determine the microvessel density. Large anastomosing sinusoidal vessels were counted as a single vessel. Large vessels with thick muscular walls were excluded from the count. The stained sections were screened at ⫻100 magnification under a light microscope to identify the five regions with the highest number of microvessels. The microvessels were counted in these areas at ⫻200 magnification (⫻20 objective and ⫻10 ocular, 0.075 mm2 per field). The microvessel count was expressed as the mean number of vessels in this area.
NUCLEAR MORPHOMETRY All the resected specimens were fixed in 4% neutral-buffered formalin and embedded routinely in paraffin wax using an automated tissue-processing unit. The histologic sections (5 m thick) were cut from each specimen, stained with hematoxylin-eosin, and assessed using an interactive imageanalysis system, consisting of a photomicroscope (Nikon, Tokyo, Japan) equipped with a high-resolution video camera (JVC, KKY-F-30, JCCD, Tokyo, Japan). For each patient, random selections of five microscope fields (⫻400) were assessed, and the nuclei of the tumor cells were outlined and measured, avoiding the nuclei of stromal cells and necrotic 905
FIGURE 1. Numerous CD31 positive microvessels are visible in the tumor tissue (CD31 immunostaining, original magnification ⫻200).
TABLE III. Statistical significance of evaluation factors for RCC stage
Urinary VEGF MVD count Nuclear area (m2)
Low Stage (T1, T2)
High Stage (T3, T4)
P Value
0.050 ⫾ 0.045 119.87 ⫾ 79.59 42.15 ⫾ 13.15
0.340 ⫾ 0.480 82.17 ⫾ 45.45 54.49 ⫾ 10.00
0.024 0.350 0.043
KEY: MVD ⫽ microvessel density; other abbreviations as in Table I.
areas. Thirty nuclei from representative fields were measured for each patient, and the mean nuclear area was calculated (in square micrometers).
STATISTICAL ANALYSIS The statistical analysis was done using the Mann-Whitney U test, analysis of variance, or Pearson’s correlation coefficient. P ⬍0.05 was considered to be statistically significant.
RESULTS The mean corrected urinary VEGF level was higher in patients with RCC than it was in the control patients. The mean corrected urinary VEGF level in patients with RCC was 0.147, but was 0.024 in the control group (P ⫽ 0.039). However, no statistically significant correlation was found between the levels of corrected urinary VEGF and age or sex (r ⫽ 0.23, P ⫽ 0.26) (Table I). Table II shows the levels of corrected urinary VEGF according to the pathologic characteristics of the RCC. The mean urinary VEGF level in highstage tumor (T3, T4) was significantly higher than in the lower stage tumor (T1, T2) (P ⫽ 0.024). However, no statistically significant correlation was found between the levels of corrected urinary 906
VEGF and the size of the tumor or nuclear area. A strong correlation between the levels of urinary VEGF and tumor cell grade was not found, but patients with grade 3 disease had significantly different levels than those with grade 1 or 2 disease, as tested by the multiple comparison test (P ⫽ 0.009). Figure 1 shows the immunohistochemical staining of CD31. Numerous positive microvessels are seen in the tumor. No statistical correlation between the microvessel density and the pathologic characteristics of the RCC, such as stage, tumor cell grade, or cell type, was found. However, an inverse correlation was noted between the microvessel density and nuclear area (r ⫽ ⫺0.549, P ⫽ 0.012). The nuclear areas were increased more in highstage RCC than in low-stage RCC (P ⫽ 0.043) and also tended to increase according to increased tumor cell grade. The urinary VEGF and nuclear areas of the tumor cells were statistically higher in the higher stage tumors than in the lower stage tumors. However, no statistically significant correlation was found with microvessel density (r ⫽ ⫺0.302) (Table III). The quantitated urinary VEGF in the RCC did not correlate with the microvessel density meaUROLOGY 58 (6), 2001
FIGURE 2. No statistical correlation was found between the urinary VEGF and microvessel density count (r ⫽ ⫺0.183, P ⫽ 0.422).
sured by immunohistochemical staining (r ⫽ ⫺0.183, P ⫽ 0.422) (Fig. 2). COMMENT Folkman13 first introduced the concept of tumor angiogenesis factors and emphasized the importance of the induction of neovascularity to sustain tumors. Clinical and experimental studies have shown that angiogenesis is a prerequisite for solid tumor growth and metastasis.14 Once a tumor has grown beyond the point at which simple diffusion can nourish it (about 106 cells), additional expansion of the tumor cell population requires the induction of new capillary vessels. Implantation of human RCC in rabbit cornea results in neovascular growth from the limbus toward the tumor implant.15 This suggests that the tumor cells have the capacity to produce growth factors or cytokines, which act in a paracrine fashion to stimulate endothelial cell growth and neovascularization. It has recently been suggested that VEGF may have an important role in the vascular biology of these tumors and in particular as a mediator of angiogenesis.16 It has been reported that increased VEGF expression is found in RCC.11,17 However, previous investigators used complicated procedures and conducted retrospective studies.17,18 Therefore, these previous studies had significant limitations for prospective clinical application. It was reported that quantification of urinary VEGF might provide a valuable noninvasive marker for the early detection of bladder tumor recurrence, as well as therapy target.12 UROLOGY 58 (6), 2001
In this study, we found that the urinary VEGF levels in patients with RCC were significantly higher than those in the controls. It has been proposed that neovascularization is necessary for continued tumor growth. However, prior studies of RCC give conflicting results. Paradis et al.19 reported that immunohistochemical staining of VEGF expression in RCC correlated positively with the grade and size of the tumor and microvascular count. Therefore, VEGF expression was a significant independent predictor of outcome and of the stage and nuclear grade. However, Slaton et al.20 reported that the expression levels of VEGF did not correlate with the microvascular density. In an animal study, regardless of the in vivo expression level of VEGF, the incidence of spontaneous lung metastasis was low, suggesting that VEGF alone was not sufficient to produce metastasis.21 Nativ et al.22 evaluated pT1 and pT2 RCC and found no correlation between the microvessel count (MVC) and tumor cell type or tumor size in RCC, but patients with RCC with a low MVC had a significantly better survival rate than did those with high-MVC RCC. Also, an inverse correlation was found between the MVC and the nuclear area.22 In our study, the urinary VEGF levels in patients with RCC were much higher than those in the normal controls and were higher in patients with highstage RCC (Stages III and IV). A strong correlation between the levels of urinary VEGF and tumor cell grade was not found, but patients with grade 3 disease had significantly different levels than did those with grade 1 or 2 disease, as tested by the multiple comparison test (P ⫽ 0.009). However, no statistically significant correlation was found between the urinary VEGF and stage, sex, tumor size, cell type, microvessel density, platelet count, or hemoglobin. An inverse correlation was noted between the microvessel density and the nuclear area (r ⫽ ⫺0.549, P ⫽ 0.012). The urinary VEGF levels did not correlate with the microvessel density of the tumor tissue even though RCC is a representative tumor of hypervascularity. This lack of correlation may have been due to several possible causes.23 However, the most viable explanation is that genetic alterations in RCC may permit their growth in an anoxic environment, making these tumors less dependent on new vessel formation. This is supported by Graeber et al.,24 who demonstrated in an animal model that p53 mutations can permit tumor growth in hypoxic areas. However, larger numbers of patients are needed to clarify this issue. CONCLUSIONS The results of this study demonstrated that urinary VEGF in patients with RCC is higher in high907
stage than it is in low-stage RCC. However, no correlation was found between urinary VEGF and microvessel density, nuclear area, or tumor size. The nuclear area was inversely correlated with the microvessel density. Therefore, in RCC, urinary VEGF levels may not reflect the underlying angiogenic activity, and it may be that other angiogenic factors play a more prominent role. REFERENCES 1. Folkman J: Tumor angiogenesis: therapeutic implications. N Engl J Med 285: 1182–1186, 1971. 2. Yoshino S, Kato M, and Okada K: Prognostic significance of microvessel counts in low stage renal cell carcinoma. Int J Urol 2: 156 –160, 1995. 3. Li VW, Folkerth RD, Watanabe H, et al: Microvessel count and cerebrospinal fluid basic fibroblast growth factor in children with brain tumors. Lancet 344: 82– 86, 1994. 4. Takahashi Y, Kitadai Y, Bucana CD, et al: Expression of vascular endothelial growth factor and its receptor, KDR, correlates with vascularity, metastasis and proliferation. Cancer Res 55: 3964 –3968, 1995. 5. Delahunt B, Bethwaite PB, and Thornton A: Prognostic significance of microscopic vascularity for clear renal cell carcinoma. Br J Urol 80: 401– 404, 1997. 6. Mancilla-Jimenez R: Papillary renal cell carcinoma: a clinical, radiologic and pathologic study of 34 cases. Cancer 38: 2469 –2480, 1976. 7. Slaton JW, Inoue K, Perrotte P, et al: Expression levels of genes that regulate metastasis and angiogenesis correlate with advanced pathological stage of renal cell carcinoma. Am J Pathol 158: 735–743, 2001. 8. Folkman J, and Klagsburn M: Angiogenic factors. Science 235: 442– 447, 1987. 9. Senger DR, Van De Water L, Brown LF, et al: Vascular permeability factor (VPF, VEGF) in tumor biology. Cancer Metastasis Rev 12: 303–324, 1993. 10. Sato K, Tsuchiya N, Sasaki R, et al: Increased serum levels of vascular endothelial growth factor in patients with renal cell carcinoma. Jpn J Cancer Res 90: 874 – 879, 1999. 11. Nicol D, Hii SI, Walsh M, et al: Vascular endothelial growth factor expression is increased in renal cell carcinoma. J Urol 157: 1482–1486, 1997. 12. Crew JP, O’Brien T, Bicknell R, et al: Urinary vascular
908
endothelial growth factor and its correlation with bladder cancer recurrence rates. J Urol 161: 799 – 804, 1999. 13. Folkman J: Anti-angiogenesis: new concept for therapy of solid tumor. Ann Surg 175: 409 – 416, 1972. 14. Horak ER, Leekm R, Klenk N, et al: Angiogenesis, assessed by platelet/endothelial cell adhesion molecule antibodies as indicator of metastases and survival in breast cancer. Lancet 340: 1120 –1124, 1992. 15. Bard RH, Mydlo JH, and Freed SZ: Detection of tumor angiogenesis factor in adenocarcinoma of kidney. Urology 27: 447– 450, 1986. 16. Takahashi A, Sasaki H, Kim SH, et al: Markedly increased amounts of messenger RNA for vascular endothelial growth factor and placenta growth factor in renal cell carcinoma associated with angiogenesis. Cancer Res 54: 4233– 4237, 1994. 17. Tomisawa M, Tokunaga T, Oshika Y, et al: Expression pattern of vascular endothelial growth factor isoform is closely correlated with tumor stage and vascularization in renal cell carcinoma. Eur J Cancer 35: 133–137, 1999. 18. Lee JS, Jung JJ, and Kim J: Quantification of angiogenesis by a computerized image analysis system in renal cell carcinoma. Anal Quant Cytol Histol 22: 469 – 474, 2000. 19. Paradis V, Lagha NB, Zeimoura L, et al: Expression of vascular endothelial growth factor in renal cell carcinoma. Virchows Arch 436: 351–356, 2000. 20. Slaton JW, Inoue K, Perrotte P, et al: Expression levels of genes that regulate metastasis and angiogenesis correlate with advance pathological stage of renal cell carcinoma. Am J Pathol 158: 735–743, 2001. 21. Kanayma H, Yano S, Kim SJ, et al: Expression of vascular endothelial growth factor by human renal cancer cells enhances angiogenesis of primary tumors and production of ascites but not metastasis to the lung in nude mice. Clin Exp Metastasis 17: 831– 840, 1999. 22. Nativ O, Sabo E, Reiss A, et al: Clinical significance of tumor angiogenesis in patients with localized renal cell carcinoma. Urology 51: 693– 696, 1998. 23. Gleich LL, Biddinger PW, Duperier FD, et al: Tumor angiogenesis as a prognostic indicator in T2–T4 oral cavity squamous cell carcinoma: a clinical–pathologic correlation. Head Neck 19: 276 –280, 1997. 24. Graeber TG, Osmanian C, Jacks T, et al: Hypoxia mediated selection of cells with diminished apoptotic potential in solid tumor. Nature 379: 88 –91, 1996.
UROLOGY 58 (6), 2001
CLINICAL SIGNIFICANCE OF URINARY VASCULAR ENDOTHELIAL GROWTH FACTOR AND MICROVESSEL DENSITY IN PATIENTS WITH RENAL CELL CARCINOMA SUNG-GOO CHANG, SUNG-HYUN JEON, SUN-JU LEE, JOONG-MYUNG CHOI, YOUN-WHA KIM
AND
ABSTRACT Objectives. To investigate the urinary vascular endothelial growth factor (VEGF) levels from patients with renal cell carcinoma (RCC). Neovascularization, an essential event for the growth of solid tumors, is regulated by a number of angiogenic factors. VEGF is thought to exert potent angiogenic activity. Methods. Urine samples were obtained before radical nephrectomy from 27 patients with RCC and 10 control subjects with no evidence of cancer or inflammatory disease. VEGF was measured by enzyme-linked immunosorbent assay in the urine and corrected according to the 24-hour urine concentration of creatinine. The microvessel density was measured by immunohistochemical staining with CD31 monoclonal antibody. Nuclear morphometry was performed by photomicroscopy. Results. The corrected urinary VEGF levels in patients with RCC were much higher than those in the normal control group (P ⫽ 0.039) and were more elevated in patients with higher stages of RCC (Stages III and IV versus Stages I and II; P ⫽ 0.024). A tendency was also noted for the VEGF levels to be higher according to cell grade. However, no statistical correlation was found between the corrected urinary VEGF and age, sex, tumor size, cell type, microvessel density, platelet count, or hemoglobin. The nuclear area was higher with more advanced-stage tumors (P ⫽ 0.043) and tended to increase according to the tumor cell grade. Conclusions. The results of this study indicate that urinary VEGF levels are increased in patients with RCC. However, they may not reflect the underlying angiogenic activity, and it may be that other angiogenic factors play a more prominent role. UROLOGY 58: 904–908, 2001. © 2001, Elsevier Science Inc.
N
eovascularization is essential to the growth and metastasis of solid tumors.1 Previous studies have shown that angiogenic factors are involved in the neovascularization of renal cell carcinoma (RCC),2 brain tumors,3 and colon cancer.4 The microvessel density in the tumor tissue correlated with the malignant potential of some solid tumors.5 RCCs are solid tumors characterized by abundant neovascularization and arteriovenous fistula formation.6 The tumor vascularity is related to the clinical outcome, because metastases are more likely in patients with highly vascularized RCC.7 From the Departments of Urology, Preventive Medicine, and Pathology, Kyung Hee University School of Medicine, Seoul, Korea Reprint requests: Sung-Goo Chang, M.D., Ph.D., Department of Urology, Kyung Hee University Medical Center, No. 1 Hoegidong Dongdaemun Ku, Seoul 130-702, Korea Submitted: May 18, 2001, accepted (with revisions): July 24, 2001
904
© 2001, ELSEVIER SCIENCE INC. ALL RIGHTS RESERVED
It has been proposed that tumor-induced angiogenesis is stimulated in a paracrine fashion by growth factors produced by tumor cells that bind to vascular endothelial cells, resulting in cellular proliferation.8 In contrast to other angiogenic growth factors, vascular endothelial growth factor (VEGF) is a secreted peptide with activity restricted to vascular endothelial cells.9 Increased serum levels of VEGF from patients with RCC have been reported,10 and these levels were also increased in RCC tissue.11 Crew et al.12 quantified VEGF in the urine of patients with bladder cancer. This study was undertaken to determine, through enzyme-linked immunosorbent assay and immunohistochemistry, the quantitation of urinary VEGF in patients with RCC and a possible relationship between the urinary VEGF levels and disease characteristics such as microvessel density, nuclear area, tumor grade, and tumor size. 0090-4295/01/$20.00 PII S0090-4295(01)01375-9
TABLE I. Patient characteristics Patients (n) Men
16
Women
11
Control
9
Corrected Urinary VEGF*
P Value
0.195 ⫾ 0.381 (0.014–1.032) 0.070 ⫾ 0.114 (0.009–1.204) 0.260
RCC
27
0.024 ⫾ 0.026 (0.000–0.040) 0.147 ⫾ 0.305 (0.009–1.204)
TABLE II. Levels of corrected urinary VEGF according to pathologic characteristics of the RCC Patients (n)
Corrected Urinary VEGF
Stage Low (T1, T2)
18
High (T3, T4)
9
0.050 ⫾ 0.045 (0.009–0.158) 0.340 ⫾ 0.480 (0.007–1.204)
P Value
0.024 0.039
KEY: VEGF ⫽ vascular endothelial growth factor; RCC ⫽ renal cell carcinoma. Data presented as the mean ⫾ SD, with the range in parentheses, unless noted otherwise. * Corrected urinary VEGF ⫽ level of urinary VEGF/24-hour urine creatinine.
Pathologic grade 1
2
2
17
3
2
0.041 ⫾ 0.048 (0.007–0.075) 0.069 ⫾ 0.098 (0.009–0.403) 0.623 ⫾ 0.821 (0.007–1.204) 0.009*
MATERIAL AND METHODS PATIENTS The patients included 16 men and 11 women (mean age 52.6 years). We analyzed 37 urine samples from 27 patients with RCC and 10 patients with ureteral stones as controls. The patients were treated with radical nephrectomy (21 cases), had a diagnosis only (4 cases), and had an angioinfarction (2 cases). The RCC stage was classified by TNM classification and the grade according to Fuhrman.
Size (cm) ⱖ7 ⬍7
13 14
0.560 Nuclear area (m2) 0–43.2 ⬎43.2
11 10
ENZYME-LINKED IMMUNOSORBENT ASSAY Urine VEGF was measured by the Quantikine (R&D System, Minneapolis, Minn) enzyme-linked immunosorbent assay. The measurable range of this method was from 15.6 to 1000 pg/mL. To control for differences in the urine concentration, urinary VEGF was expressed relative to the 24-hour urinary creatinine content.
IMMUNOHISTOCHEMISTRY Immunohistochemical staining was performed using a streptavidin-biotin immunoperoxidase method according to the supplier’s protocol (Dako, LSAB Kit, Carpenteria, Calif). In brief, paraffin-embedded sections were deparaffinized in xylene and rehydrated with graded ethanol. After quenching the endogenous peroxidase activity in 0.3% hydrogen peroxide for 30 minutes and blocking reagents for 30 minutes, primary monoclonal mouse anti-human endothelial cell, CD31 (Dako), was applied to the sections at a dilution of 1:20 and incubated in a moist chamber for 2 hours at room temperature. After washing out the excess complex, the localization of antibodies was visualized by incubating the section for 10 minutes in 3,3⬘-diaminobenzidine tetrahydrochloride (Research Genetics, Huntville, Ala). The positive controls for human erythrocyte glucose transporter (Glut-1) were the red blood cells present in each section. The negative control sections were made of horse serum without the primary Glut-1 UROLOGY 58 (6), 2001
0.038 ⫾ 0.034 (0.007–0.110) 0.200 ⫾ 0.369 (0.009–1.204) 0.205
URINE COLLECTION A midstream urine specimen (10 mL) was collected from patients before treatment, and the creatinine concentration in the 24-hour collected urine specimen was measured by the Jaffe reaction method. The collected urine for measurement of VEGF was centrifuged at 3000g for 3 minutes. The supernatant was separated and stored at ⫺70°C until testing.
0.058 ⫾ 0.048 (0.009–1.204) 0.042 ⫾ 0.043 (0.014–0.770)
KEY: Abbreviations as in Table I. Data presented as the mean ⫾ SD, with the range in parentheses, unless noted otherwise. * Tested by analysis of variance.
antibody. One pathologist who was unaware of the pathologic findings independently interpreted the stained sections. Vessels with a clearly defined lumen or well-defined linear vessel shape that stained brown, but not single endothelial cells, were counted to determine the microvessel density. Large anastomosing sinusoidal vessels were counted as a single vessel. Large vessels with thick muscular walls were excluded from the count. The stained sections were screened at ⫻100 magnification under a light microscope to identify the five regions with the highest number of microvessels. The microvessels were counted in these areas at ⫻200 magnification (⫻20 objective and ⫻10 ocular, 0.075 mm2 per field). The microvessel count was expressed as the mean number of vessels in this area.
NUCLEAR MORPHOMETRY All the resected specimens were fixed in 4% neutral-buffered formalin and embedded routinely in paraffin wax using an automated tissue-processing unit. The histologic sections (5 m thick) were cut from each specimen, stained with hematoxylin-eosin, and assessed using an interactive imageanalysis system, consisting of a photomicroscope (Nikon, Tokyo, Japan) equipped with a high-resolution video camera (JVC, KKY-F-30, JCCD, Tokyo, Japan). For each patient, random selections of five microscope fields (⫻400) were assessed, and the nuclei of the tumor cells were outlined and measured, avoiding the nuclei of stromal cells and necrotic 905
FIGURE 1. Numerous CD31 positive microvessels are visible in the tumor tissue (CD31 immunostaining, original magnification ⫻200).
TABLE III. Statistical significance of evaluation factors for RCC stage
Urinary VEGF MVD count Nuclear area (m2)
Low Stage (T1, T2)
High Stage (T3, T4)
P Value
0.050 ⫾ 0.045 119.87 ⫾ 79.59 42.15 ⫾ 13.15
0.340 ⫾ 0.480 82.17 ⫾ 45.45 54.49 ⫾ 10.00
0.024 0.350 0.043
KEY: MVD ⫽ microvessel density; other abbreviations as in Table I.
areas. Thirty nuclei from representative fields were measured for each patient, and the mean nuclear area was calculated (in square micrometers).
STATISTICAL ANALYSIS The statistical analysis was done using the Mann-Whitney U test, analysis of variance, or Pearson’s correlation coefficient. P ⬍0.05 was considered to be statistically significant.
RESULTS The mean corrected urinary VEGF level was higher in patients with RCC than it was in the control patients. The mean corrected urinary VEGF level in patients with RCC was 0.147, but was 0.024 in the control group (P ⫽ 0.039). However, no statistically significant correlation was found between the levels of corrected urinary VEGF and age or sex (r ⫽ 0.23, P ⫽ 0.26) (Table I). Table II shows the levels of corrected urinary VEGF according to the pathologic characteristics of the RCC. The mean urinary VEGF level in highstage tumor (T3, T4) was significantly higher than in the lower stage tumor (T1, T2) (P ⫽ 0.024). However, no statistically significant correlation was found between the levels of corrected urinary 906
VEGF and the size of the tumor or nuclear area. A strong correlation between the levels of urinary VEGF and tumor cell grade was not found, but patients with grade 3 disease had significantly different levels than those with grade 1 or 2 disease, as tested by the multiple comparison test (P ⫽ 0.009). Figure 1 shows the immunohistochemical staining of CD31. Numerous positive microvessels are seen in the tumor. No statistical correlation between the microvessel density and the pathologic characteristics of the RCC, such as stage, tumor cell grade, or cell type, was found. However, an inverse correlation was noted between the microvessel density and nuclear area (r ⫽ ⫺0.549, P ⫽ 0.012). The nuclear areas were increased more in highstage RCC than in low-stage RCC (P ⫽ 0.043) and also tended to increase according to increased tumor cell grade. The urinary VEGF and nuclear areas of the tumor cells were statistically higher in the higher stage tumors than in the lower stage tumors. However, no statistically significant correlation was found with microvessel density (r ⫽ ⫺0.302) (Table III). The quantitated urinary VEGF in the RCC did not correlate with the microvessel density meaUROLOGY 58 (6), 2001
FIGURE 2. No statistical correlation was found between the urinary VEGF and microvessel density count (r ⫽ ⫺0.183, P ⫽ 0.422).
sured by immunohistochemical staining (r ⫽ ⫺0.183, P ⫽ 0.422) (Fig. 2). COMMENT Folkman13 first introduced the concept of tumor angiogenesis factors and emphasized the importance of the induction of neovascularity to sustain tumors. Clinical and experimental studies have shown that angiogenesis is a prerequisite for solid tumor growth and metastasis.14 Once a tumor has grown beyond the point at which simple diffusion can nourish it (about 106 cells), additional expansion of the tumor cell population requires the induction of new capillary vessels. Implantation of human RCC in rabbit cornea results in neovascular growth from the limbus toward the tumor implant.15 This suggests that the tumor cells have the capacity to produce growth factors or cytokines, which act in a paracrine fashion to stimulate endothelial cell growth and neovascularization. It has recently been suggested that VEGF may have an important role in the vascular biology of these tumors and in particular as a mediator of angiogenesis.16 It has been reported that increased VEGF expression is found in RCC.11,17 However, previous investigators used complicated procedures and conducted retrospective studies.17,18 Therefore, these previous studies had significant limitations for prospective clinical application. It was reported that quantification of urinary VEGF might provide a valuable noninvasive marker for the early detection of bladder tumor recurrence, as well as therapy target.12 UROLOGY 58 (6), 2001
In this study, we found that the urinary VEGF levels in patients with RCC were significantly higher than those in the controls. It has been proposed that neovascularization is necessary for continued tumor growth. However, prior studies of RCC give conflicting results. Paradis et al.19 reported that immunohistochemical staining of VEGF expression in RCC correlated positively with the grade and size of the tumor and microvascular count. Therefore, VEGF expression was a significant independent predictor of outcome and of the stage and nuclear grade. However, Slaton et al.20 reported that the expression levels of VEGF did not correlate with the microvascular density. In an animal study, regardless of the in vivo expression level of VEGF, the incidence of spontaneous lung metastasis was low, suggesting that VEGF alone was not sufficient to produce metastasis.21 Nativ et al.22 evaluated pT1 and pT2 RCC and found no correlation between the microvessel count (MVC) and tumor cell type or tumor size in RCC, but patients with RCC with a low MVC had a significantly better survival rate than did those with high-MVC RCC. Also, an inverse correlation was found between the MVC and the nuclear area.22 In our study, the urinary VEGF levels in patients with RCC were much higher than those in the normal controls and were higher in patients with highstage RCC (Stages III and IV). A strong correlation between the levels of urinary VEGF and tumor cell grade was not found, but patients with grade 3 disease had significantly different levels than did those with grade 1 or 2 disease, as tested by the multiple comparison test (P ⫽ 0.009). However, no statistically significant correlation was found between the urinary VEGF and stage, sex, tumor size, cell type, microvessel density, platelet count, or hemoglobin. An inverse correlation was noted between the microvessel density and the nuclear area (r ⫽ ⫺0.549, P ⫽ 0.012). The urinary VEGF levels did not correlate with the microvessel density of the tumor tissue even though RCC is a representative tumor of hypervascularity. This lack of correlation may have been due to several possible causes.23 However, the most viable explanation is that genetic alterations in RCC may permit their growth in an anoxic environment, making these tumors less dependent on new vessel formation. This is supported by Graeber et al.,24 who demonstrated in an animal model that p53 mutations can permit tumor growth in hypoxic areas. However, larger numbers of patients are needed to clarify this issue. CONCLUSIONS The results of this study demonstrated that urinary VEGF in patients with RCC is higher in high907
stage than it is in low-stage RCC. However, no correlation was found between urinary VEGF and microvessel density, nuclear area, or tumor size. The nuclear area was inversely correlated with the microvessel density. Therefore, in RCC, urinary VEGF levels may not reflect the underlying angiogenic activity, and it may be that other angiogenic factors play a more prominent role. REFERENCES 1. Folkman J: Tumor angiogenesis: therapeutic implications. N Engl J Med 285: 1182–1186, 1971. 2. Yoshino S, Kato M, and Okada K: Prognostic significance of microvessel counts in low stage renal cell carcinoma. Int J Urol 2: 156 –160, 1995. 3. Li VW, Folkerth RD, Watanabe H, et al: Microvessel count and cerebrospinal fluid basic fibroblast growth factor in children with brain tumors. Lancet 344: 82– 86, 1994. 4. Takahashi Y, Kitadai Y, Bucana CD, et al: Expression of vascular endothelial growth factor and its receptor, KDR, correlates with vascularity, metastasis and proliferation. Cancer Res 55: 3964 –3968, 1995. 5. Delahunt B, Bethwaite PB, and Thornton A: Prognostic significance of microscopic vascularity for clear renal cell carcinoma. Br J Urol 80: 401– 404, 1997. 6. Mancilla-Jimenez R: Papillary renal cell carcinoma: a clinical, radiologic and pathologic study of 34 cases. Cancer 38: 2469 –2480, 1976. 7. Slaton JW, Inoue K, Perrotte P, et al: Expression levels of genes that regulate metastasis and angiogenesis correlate with advanced pathological stage of renal cell carcinoma. Am J Pathol 158: 735–743, 2001. 8. Folkman J, and Klagsburn M: Angiogenic factors. Science 235: 442– 447, 1987. 9. Senger DR, Van De Water L, Brown LF, et al: Vascular permeability factor (VPF, VEGF) in tumor biology. Cancer Metastasis Rev 12: 303–324, 1993. 10. Sato K, Tsuchiya N, Sasaki R, et al: Increased serum levels of vascular endothelial growth factor in patients with renal cell carcinoma. Jpn J Cancer Res 90: 874 – 879, 1999. 11. Nicol D, Hii SI, Walsh M, et al: Vascular endothelial growth factor expression is increased in renal cell carcinoma. J Urol 157: 1482–1486, 1997. 12. Crew JP, O’Brien T, Bicknell R, et al: Urinary vascular
908
endothelial growth factor and its correlation with bladder cancer recurrence rates. J Urol 161: 799 – 804, 1999. 13. Folkman J: Anti-angiogenesis: new concept for therapy of solid tumor. Ann Surg 175: 409 – 416, 1972. 14. Horak ER, Leekm R, Klenk N, et al: Angiogenesis, assessed by platelet/endothelial cell adhesion molecule antibodies as indicator of metastases and survival in breast cancer. Lancet 340: 1120 –1124, 1992. 15. Bard RH, Mydlo JH, and Freed SZ: Detection of tumor angiogenesis factor in adenocarcinoma of kidney. Urology 27: 447– 450, 1986. 16. Takahashi A, Sasaki H, Kim SH, et al: Markedly increased amounts of messenger RNA for vascular endothelial growth factor and placenta growth factor in renal cell carcinoma associated with angiogenesis. Cancer Res 54: 4233– 4237, 1994. 17. Tomisawa M, Tokunaga T, Oshika Y, et al: Expression pattern of vascular endothelial growth factor isoform is closely correlated with tumor stage and vascularization in renal cell carcinoma. Eur J Cancer 35: 133–137, 1999. 18. Lee JS, Jung JJ, and Kim J: Quantification of angiogenesis by a computerized image analysis system in renal cell carcinoma. Anal Quant Cytol Histol 22: 469 – 474, 2000. 19. Paradis V, Lagha NB, Zeimoura L, et al: Expression of vascular endothelial growth factor in renal cell carcinoma. Virchows Arch 436: 351–356, 2000. 20. Slaton JW, Inoue K, Perrotte P, et al: Expression levels of genes that regulate metastasis and angiogenesis correlate with advance pathological stage of renal cell carcinoma. Am J Pathol 158: 735–743, 2001. 21. Kanayma H, Yano S, Kim SJ, et al: Expression of vascular endothelial growth factor by human renal cancer cells enhances angiogenesis of primary tumors and production of ascites but not metastasis to the lung in nude mice. Clin Exp Metastasis 17: 831– 840, 1999. 22. Nativ O, Sabo E, Reiss A, et al: Clinical significance of tumor angiogenesis in patients with localized renal cell carcinoma. Urology 51: 693– 696, 1998. 23. Gleich LL, Biddinger PW, Duperier FD, et al: Tumor angiogenesis as a prognostic indicator in T2–T4 oral cavity squamous cell carcinoma: a clinical–pathologic correlation. Head Neck 19: 276 –280, 1997. 24. Graeber TG, Osmanian C, Jacks T, et al: Hypoxia mediated selection of cells with diminished apoptotic potential in solid tumor. Nature 379: 88 –91, 1996.
UROLOGY 58 (6), 2001