The Relationship of Vascular Endothelial Growth Factor and Coagulation Factor (Fibrin and Fibrinogen) Expression in Clear Cell Renal Cell Carcinoma

Oncology The Relationship of Vascular Endothelial Growth Factor and Coagulation Factor (Fibrin and Fibrinogen) Expression in Clear Cell Renal Cell Carcinoma Henk M. W. Verheul, Karen van Erp, Marjolein Y. V. Homs, G. S. Yoon, Petra van Der Groep, Craig Rogers, Donna E. Hansel, George J. Netto, and Roberto Pili OBJECTIVES

METHODS

RESULTS

CONCLUSIONS

To investigate the relationship between angiogenesis and coagulation markers in tumor tissues of primary renal cell carcinoma (RCC). Tumors stimulate angiogenesis and activate the coagulation cascade. The importance of the interplay between these pathways for RCC is unknown. In all, 69 clear cell RCC specimens were analyzed by immunohistochemical staining applied to tissue microarrays. The expression of vascular endothelial growth factor (VEGF), hypoxiainducible factor-1␣, fibrinogen and fibrin, and microvessel density were visually analyzed. Finally, staining patterns were related to clinical variables and survival. The VEGF expression was detected in 100% of tumors, with 68% showing a high expression, whereas hypoxia-inducible factor-1␣ staining was low (only 26% had a moderate to high staining). Fibrinogen was expressed adjacent to the tumor cells in 26% of cases, whereas in 84% it was expressed around the blood vessels. In 30% of tumors, expression of fibrin was detected. High tumor VEGF expression correlated with high fibrin staining (P ⫽ .05). From a multivariate analysis, microvessel density (P ⫽ .033) and fibrinogen adjacent to tumor cells (P ⫽ .046) were independent factors related to VEGF expression. In this study, we found clinical evidence for the permeability activity of VEGF as reflected by extravascular fibrinogen expression adjacent to tumor cells in the extracellular matrix. In addition, VEGF and fibrin expression were associated, indicative for concomitant activation of the coagulation cascade and angiogenesis in RCC. Taken together, these data indicate that activation of angiogenesis and coagulation are related in RCC. UROLOGY 75: 608 – 614, 2010. © 2010 Elsevier Inc.

A

ngiogenesis is essential for tumor development and metastasis formation. Vascular endothelial growth factor (VEGF) plays a major role in stimulating new blood vessel formation in several biological processes including tumor angiogenesis. Tumors cannot grow beyond the critical size of 1-2 mm (or about 106 cells) without angiogenesis because of the limited diffusion of oxygen and nutrients.1 An important factor in the switch to an angiogenic phenotype is the increase in hypoxia in the microenvironment of a grow-

H.M.W. Verheul and K. van Erp contributed equally to this work. From the Department of Medical Oncology, VU Medical Center, Amsterdam, The Netherlands; Department of Oncology, The Johns Hopkins University, Baltimore, Maryland; Department of Medical Oncology, University Medical Center Utrecht, Utrecht, The Netherlands; Department of Pathology, The Johns Hopkins University, Baltimore, Maryland; Department of Pathology, University Medical Center Utrecht, Utrecht, The Netherlands; and Department of Urology, The Johns Hopkins University, Baltimore, Maryland Reprint requests: Roberto Pili, M.D., Roswell Park Cancer Institute, Elm & Carlton St, Buffalo, MD 21231. E-mail: [email protected] Submitted: December 6, 2008, accepted (with revisions): May 30, 2009

608

© 2010 Elsevier Inc. All Rights Reserved

ing microscopic tumor. Hypoxia-inducible factor (HIF) plays an essential role in the O2 homeostasis.2 During hypoxia, HIF-␣ binds to HIF-␤. Subsequently, the HIF ␣/␤ heterodimer activates genes that regulate adaptation to hypoxia by increasing VEGF production and stimulating angiogenesis.3 The formation of tumor vasculature has been extensively studied. Tumor blood vessels are structurally and functionally irregular, with dead ends, disordered blood flow, and increased permeability.1 VEGF plays a crucial role in the formation of new vessels.4 High VEGF expression in renal cell carcinoma (RCC) tissues has been shown previously.5 Tumor vascularity, assessed histologically as microvessel density (MVD), has been suggested as a potential prognosticator. In lung cancer, highly vascularized tumor, reflected by a high MVD, appears to be associated with higher likelihood for metastasis and poor prognosis.6 However, in contrast to initial expectations based on other tumor types and the importance of the VHL mutation in renal tumors pathogenesis, a high MVD does 0090-4295/10/$34.00 doi:10.1016/j.urology.2009.05.075

not correlate with poor prognosis in RCCs.7-10 This finding might well be due to the type of tumor vessels that are being analyzed. In a study of Yao et al,10 a difference in prognostic value between undifferentiated and more differentiated microvessel densities was observed. In addition, in RCC the relationship between the expression of the angiogenic factor, VEGF, and MVD has not been found.5,11,12 This lack of a relationship might be due to the presence of pre-existent kidney tissue that is highly vascularized. However, the importance of angiogenesis for RCC is reflected by the efficacy of antiangiogenic therapy. In the past 2 years, a number of antiangiogenic agents have shown clinical activity in RCC.13-16 The clinical development of these angiogenesis inhibitors has been complicated by new types of side effects. The most common side effects are hypertension, fatigue, skin toxicities, gastrointestinal perforations, thrombotic events, bleeding, and impaired wound healing. Although the underlying mechanisms are not completely understood, some of these side effects are related to the interaction between angiogenesis and the hemostatic system.17 Activation of the coagulation cascade results in stimulation of angiogenesis and vice versa. Several angiogenic factors stimulate activation of the coagulation cascade.18 For example, VEGF stimulates tissue factor (TF) expression on endothelial cells. TF is the main regulator of the coagulation cascade by inducing thrombin formation, which in turn activates platelets and converts fibrinogen into fibrin, resulting in clot formation.17 Fibrin can be assessed as a marker for activation of the coagulation cascade. Activation of the coagulation cascade has been recognized in several types of cancer. It has been suggested that activation of the coagulation cascade may contribute to the tumor biology of RCC. Clinical evidence of such an interaction has been seen in patients with intravascular tumor thrombi formations indicative for the hypercoagulable state of this tumor type. The aim of this retrospective study was to investigate the importance of the concomitant activation of angiogenesis and the coagulation cascade in RCC. Expression of angiogenic markers (VEGF, MVD, and HIF) and coagulation factors fibrinogen and fibrin in resected RCC lesions was assessed by immunohistochemistry. The relationship between the expression patterns was studied and the expression was related to the prognosis of these patients with RCC, although the design of this retrospective study was not powered for prognostic relations.

MATERIAL AND METHODS The study included 69 patients with clear cell RCC who underwent surgery between May 1984 and July 2002 at the Johns Hopkins Hospital, Baltimore, MD. Surgical specimens were analyzed and stored in the Pathology Department of the Johns Hopkins University. Patient and tumor characteristics are shown in Table 1. UROLOGY 75 (3), 2010

Table 1. Patient and tumor characteristics of 69 patients with clear cell renal cell carcinoma N (%) Age (yr ⫾ SD) Stage I II III IV Surgery Radical nephrectomy Partial nephrectomy Resection metastasis Tumor size (cm ⫾ SD) Fuhrman nuclear grade 2 3 4

61.1 ⫾ 11.4 Range 28-87 10 (16) 4 (6) 43 (68) 6 (10) 61 (95) 2 (3) 1 (2) 8.1 ⫾ 2.8 Range 3-15 30 (44) 33 (48) 6 (9)

Immunohistochemistry For tissue microarray (TMA), representative paraffin tumor blocks were selected by primary evaluation of hematoxylineosin-stained slides before TMA preparation. Normal kidney tissue adjacent to the tumor in all specimens was used for internal control. From each specimen, 4 tumor and 4 benign tissue “spots” were represented on the TMA. For immunohistochemical analysis, 4-␮m thick TMA sections were deparaffinized and rehydrated. Epitope retrieval was performed with citrate buffer in a steamer for 30 minutes, with the exception of fibrinogen staining where proteinase K (Invitrogen, Carlsbad, CA) was used at 37°C for 5 minutes. Endogenous peroxidase activity was blocked for 30 minutes in methanol containing 0.3% hydrogen peroxidase. Thereafter the catalyzed signal amplification (CSA) system (DAKO, Glostrup, Denmark) was used for HIF-1␣ staining with the mouse monoclonal antibody (Abcam, Cambridge, UK) according to the manufacturer’s instruction. The other antibodies were detected by the avidin– biotin complex (ABC) method with the VECTASTAIN ABC Kit (Vector laboratories, Burlingame, CA). Slides were incubated overnight at 4°C with one of the following primary antibodies: mouse-anti-human VEGF 1:500 (R&D Systems, Minneapolis, MN), mouse-anti-human CD31 1:200 (Dako), rabbit-anti-human fibrinogen 1:500 (Dako), mouseanti-human fibrin 1:250 (Accurate Chem and Science, Corp, Westbury, NY). Stainings were developed with diaminobenzidine. Appropriate negative controls (obtained using the IgG isotype controls and omission of the primary antibody) and positive controls were used throughout the staining procedures. Immunohistochemical staining scoring methods were designed by 3 investigators (H.V., K.v.E., and G.N.) based on previously published methods and staining patterns. The TMAs were independently scored for staining patterns by 2 independent investigators (H.V. and K.v.E.) who were unaware of clinicopathologic variables. For VEGF, staining of the tumor cells and the vasculature were separately scored and the pattern of staining was described (focal, multifocal, or diffuse). A semiquantative scoring system was used: ⫺, no staining; ⫹, mild staining; ⫹⫹, moderate staining; and ⫹⫹⫹, strong staining. These were then categorized as low staining (⫺/⫹) and high staining (⫹⫹/⫹⫹⫹). Microvessel density was determined at 200⫻ magnification, counting the whole tissue core on the TMA, excluding areas with prominent necrosis. For HIF-1␣, 609

the immunochemical results were classified as follows: (no staining), ⫹ (nuclear staining in ⬍ 10% of cells), ⫹⫹ (10%30% staining), ⫹⫹⫹ (30%-100% staining) and were then categorized as low staining (⫺/⫹) and high staining (⫹⫹/ ⫹⫹⫹). For fibrinogen, the staining matrix adjacent to tumor cells and the staining around the blood vessels were separately scored similar to the VEGF staining. For fibrin, the intensity of staining was described (⫺ [none], ⫹ [⬍ 5%], ⫹⫹ [5%-25%], and ⫹⫹⫹ [⬎ 25%]) and categorized as no staining and positive staining and whether the staining was located mainly intravascularly or extravascularly.

Statistical Analysis Immunohistochemistry results were correlated with clinical and pathologic variables and survival. For each tumor, average values for VEGF and HIF-1␣, MVD, fibrin, and fibrinogen were calculated from all informative TMA spots and categorized as described previously. The relationship between VEGF, HIF-1␣, fibrinogen, and fibrin were analyzed using ␹2 test. MVD in relation to these markers was analyzed with the nonparametric Mann–Whitney test and correlations were calculated with Spearman correlation coefficient. A multivariate logistic analysis was performed including VEGF as a dependent factor and MVD, HIF-1␣, fibrinogen staining adjacent to tumor cells, fibrinogen adjacent to vessels, fibrin, tumor stage, tumor size and Fuhrman nuclear grade as independent factors. First, a univariate analysis was performed, and then variables with significance ⬍ 0.20 were selected for the multivariate analysis. Survival rates were calculated using the Kaplan–Meier method. Differences in survival for the different markers or patient and tumor characteristics were analyzed using log-rank test. P ⬍.05 was considered statistically significant. All statistical analyses were performed with the statistical software SPSS version 15.0.

RESULTS Angiogenic Markers The expression of VEGF was detected in 100% of the tumors, with 68% showing a high expression (Fig. 1A). Staining patterns were mainly multifocal or diffuse (Table 2). VEGF staining of the vessels was slightly lower, with 96% showing expression including 56% high expression. The mean MVD was 54 vessels per counted tissue core with a standard deviation of 35 (range 8-161) (Fig. 1B). HIF-1␣ expression was low in 74% of patients; only 26% of patients had moderate to high staining (Fig. 1C). Staining was localized to the nuclei of the cells. Coagulation Markers High fibrinogen staining adjacent to tumor cells in the extracellular matrix was seen in 18 patient samples (26%) compared with staining adjacent to blood vessels in 58 patient samples (84%) (Fig. 1D). Activation of the coagulation cascade was found only in a minority of the tumors as documented by the low fibrin-staining rates. Forty-eight specimens (70%) showed no expression, 13 (19%) showed a focal expression, and only 8 (12%) showed a multifocal or diffuse expression (Fig. 1E). The localization was intravascular in 62% (n ⫽ 13) and extravascular in 38% (n ⫽ 8). 610

Relationship Between Angiogenic Factors and Coagulation Factors A higher VEGF staining of the tumor correlated with a lower MVD (P ⫽ .05) (Table 3). No association was observed between VEGF and HIF-1␣ staining, neither for HIF-1␣ staining and MVD (Table 3). High VEGF staining was associated with high fibrin staining (P ⫽ .05). From the univariate analysis, MVD, fibrinogen adjacent to tumor cells, and fibrinogen adjacent to vessels were selected for multivariate analysis (P ⬍.20). From this multivariate analysis, MVD (P ⫽ .033) and fibrinogen adjacent to tumor cells (P ⫽ .046) were independent factors related to VEGF expression. Prognostic Factors Complete follow-up data were available for 61 patients, of which 38 patients had died. Mean survival was 7.7 years (95% confidence interval 6.2-9.1). A high stage of disease (P ⫽ .029) and a higher Fuhrman nuclear grade (P ⫽ .049) were associated with decreased survival (Fig. 2). Although overall, patients with higher tumor VEGF expression appeared to have a worse rate of survival, the relationship between VEGF expression and survival was not statistically significant (P ⫽ .13), (Fig. 2). None of the other markers were of prognostic significance.

COMMENT This study indicates that activation of angiogenesis and coagulation is present in RCC. In addition, this is the first study in patients, showing that the vascular permeability activity of VEGF is reflected by fibrinogen extravasation. The association between VEGF and fibrin expression indicates concomitant activation of the coagulation cascade and angiogenesis. Second, this study provides evidence that in RCC, higher VEGF expression is not associated with higher MVD, and it even suggests that there may be an inverse relationship between the 2 parameters. VEGF expression was related to fibrinogen staining adjacent to tumor cells in the extracellular matrix. VEGF, also known as the vascular permeability factor, can induce leakage of plasma proteins from the intravascular compartment into the extravascular compartment by activation of the tumor vascular cells. Therefore, the expression of the plasma protein fibrinogen is increased in tumors that express high VEGF levels. Extravasated fibrinogen can be converted by thrombin in the local tumor microenvironment into fibrin.18 In this study, we found that high VEGF expression was related to high fibrin expression. Fibrin was expressed in 30% of the tumor tissues, whereas fibrinogen expression adjacent to tumor cells was abundantly expressed (64% showed mild staining and 26%, moderate to high staining). Fibrin stabilizes platelet aggregates at the sites of vascular injury, thus stabilizing hemostatic plugs. Fibrin is an activation marker of the coagulation cascade. The fibrin UROLOGY 75 (3), 2010

A

B

C

D

E

Figure 1. VEGF, MVD, HIF-1␣, fibrin and fibrinogen expression in renal cell carcinoma. (A) VEGF: left, low staining; right, high staining, (B) MVD: left, low MVD, arrows point to vessels; right, high MVD, (C) HIF-1␣: left, low staining; right, high staining, (D) Fibrinogen: left, low staining; right, high staining adjacent to tumor cells, (E) Fibrin: left, low staining; right, high staining.

UROLOGY 75 (3), 2010

611

Table 2. VEGF staining of TMA in 69 surgical specimens of clear cell renal cell carcinomas VEGF Tumor staining (%) ⫺ ⫹ ⫹⫹ ⫹⫹⫹ (Adjacent to) vessel staining (%) ⫺ ⫹ ⫹⫹ ⫹⫹⫹

HIF-1␣ Fibrinogen

0 (0) 10 (15) 22 (32) 38 (59) 32 (47) 10 (15) 14 (21) 7 (11)

7 (10) 44 (64) 17 (25) 1 (1)

3 (4) 27 (40) 31 (46) 7 (10)

0 (0) 11 (16) 46 (67) 12 (17)

Fibrin 48 (70) 13 (19) 4 (6) 4 (6)

Table 3. VEGF staining in relation to MVD, HIF-1␣, and coagulation markers fibrinogen and fibrin

MVD* Mean HIF Low High Fibrinogen Adjacent to tumor cells Low High Fibrinogen Adjacent to vessels Low High Fibrin No staining Staining

VEGF Low

VEGF High

P

68

47

.052

14 8

34 9

.18

19 3

31 15

.10

4 18

7 39

⬎ .20

19 3

29 17

.048*

* Spearman correlation coefficient VEGF staining with mean MVD: r ⫽ ⫺0.26 P ⫽ .04.

matrix is essential for physiological and pathologic events, including inflammation, hemostasis, wound healing, and tumor angiogenesis. Fibrin may provide a provisional proangiogenic scaffold that supports vessel formation and stimulates endothelial cell proliferation during tumor development. One explanation for the low fibrin staining in our TMA stainings might be that the activation of the coagulation cascade is relatively low. Alternatively, the expression of fibrin may be low due to the activity of plasmin. In the cascade of wound healing and repair, fibrinolysis plays an essential role to regain the homeostasis of the endothelial cell lining of the vasculature. Plasmin is generated by the conversion of plasminogen into plasmin by either urokinase plasminogen activator (uPA) or tissue-type plasminogen activator on endothelial or other cells, including tumor cells. Plasmin breaks down fibrin into fibrin-degradation products. We did not analyze the proteins involved in the fibrinolysis pathway such as uPA or tissue-type plasminogen activator expression or plasmin activity. However, previous studies have detected a high expression of uPa in RCC, 612

Figure 2. Survival curves. (A) Survival curve according to stage of disease divided into stages 1/2 and 3/4. (B) Survival curve according to Fuhrman nuclear grade divided into stages 2 and 3/4. (C) Survival curve according to VEGF expression in renal cell carcinoma.

supporting the latter hypothesis, and a high uPA was correlated with metastasis formation.19 Fibrinolysis might be regulated by the VHL pathway as well.20 Therefore, in future experiments the expression of TF, uPA, and plasmin should be considered. In our study, only paraffinembedded TMA tissue was available, whereas frozen tissue sections are required for accurate analysis of TF, uPA, and plasmin. Currently, we are developing a TMA of frozen tissue sections of RCC and are planning to perform these additional stainings. The inverse relation between VEGF expression and MVD, and the lack of a relationship between MVD and survival or prognostic factors (Fuhrman nuclear grade, tumor stage) is surprising, but it may be specific to RCC. UROLOGY 75 (3), 2010

VEGF plays an important role in the development and maintenance of the early vasculature in the healthy kidney. VEGF remains continuously expressed in a normal adult kidney in comparison with other organs. It has been hypothesized that VEGF plays a specific functional role in the glomerulus by regulating vascular permeability.21 The inverse relationship between VEGF expression and MVD might exist because VEGF is continuously expressed in kidneys. Yao et al described 2 types of endothelial cells in RCC, differentiated (CD34⫹) and undifferentiated (CD31⫹/CD34⫺), lining the vasculature. CD31 is expressed in both differentiated and undifferentiated endothelial cells. These investigators found that a higher differentiated MVD was correlated with longer survival and a higher undifferentiated MVD was correlated with a relatively shorter survival.10 Previous studies showing an inverse relationship between MVD and survival mostly analyzed differentiated vessels (CD34⫹).7-9 CD31 expression represents both and did not show a relationship with survival in the study of Yao et al,10 which we confirmed in this study. Further studies are needed to explore the relationship between VEGF and differentiated and undifferentiated MVD. It might be that the newly formed vessels due to VEGF tumor expression are less differentiated, and therefore less organized compared with the high number of differentiated vessels in the healthy kidney, resulting in a reduced MVD. HIF-1␣ expression was relatively low in this study and was not related to VEGF expression or survival. HIF-␣ family members include HIF-1␣, HIF-2␣, and HIF-3␣. HIF-1␣ and HIF-2␣ are the main factors involved in hypoxia-driven binding to HIF-␤. In vitro and in vivo studies revealed that with loss of the tumor suppressor gene VHL, HIF-2␣ might be the most important HIF isoform associated with angiogenesis. HIF-1␣ expression in VHL-mutated tumors might be proapoptotic and can be even antiproliferative. HIF-1␣ seems to have tumorpromoting and tumor-inhibiting functions depending on the cell type.22,23 Only a few studies have described immunohistochemical staining of HIF-1␣ in human RCC. Zhong et al2 showed first in their study overexpression of HIF-1␣ in different tumor types and a higher expression in metastatic breast cancer. Only 1 RCC was included in that study. A small study revealed in 73% (8/11) of the tumors an elevated expression of HIF-1␣ in sporadic RCC24 and another small study showed in 83% (10/12) of tumors an elevated expression of HIF-1␣ in patients with VHL disease.25 In a study with 31 RCC tissues, 40% of tumors had high expression levels of HIF-1␣, which was correlated with VEGF expression.26 The only study performed with a high number of patients (216 patients with RCC) revealed that 57% of tumors (123/216) had a high HIF-1␣ expression level.27 In this study, patients with a high HIF-1␣ expression tended to have a better prognosis. Taken together, HIF-1␣ expression might be less important compared with HIF-2␣ expression in human RCC with regard to survival. UROLOGY 75 (3), 2010

In conclusion, this study confirms the close relationship between angiogenesis and coagulation in cancer. In addition, we found clinical evidence for the vascular permeability activity of VEGF by assessing fibrinogen extravasation. The direct relationship between VEGF and fibrin expression reflects the concomitant activation of the coagulation cascade and angiogenesis in primary RCC. This association between coagulation and angiogenesis provides also the rationale for combination studies with selective inhibitors in the treatment of RCC. References 1. Carmeliet P. VEGF as a key mediator of angiogenesis in cancer. Oncology. 2005;69(Suppl 3):4-10. 2. Zhong H, De Marzo AM, Laughner E, et al. Overexpression of hypoxia-inducible factor 1alpha in common human cancers and their metastases. Cancer Res. 1999;59:5830-5835. 3. Kaelin WG Jr. The von Hippel–Lindau tumor suppressor protein and clear cell renal carcinoma. Clin Cancer Res. 2007;13:680s-684s. 4. Hicklin DJ, Ellis LM. Role of the vascular endothelial growth factor pathway in tumor growth and angiogenesis. J Clin Oncol. 2005;23: 1011-1027. 5. Zhang X, Yamashita M, Uetsuki H, et al. Angiogenesis in renal cell carcinoma: evaluation of microvessel density, vascular endothelial growth factor and matrix metalloproteinases. Int J Urol. 2002;9: 509-514. 6. Macchiarini P, Fontanini G, Hardin MJ, et al. Relation of neovascularisation to metastasis of non-small-cell lung cancer. Lancet. 1992;340:145-146. 7. Imao T, Egawa M, Takashima H, et al. Inverse correlation of microvessel density with metastasis and prognosis in renal cell carcinoma. Int J Urol. 2004;11:948-953. 8. Rioux-Leclercq N, Epstein JI, Bansard JY, et al. Clinical significance of cell proliferation, microvessel density, and CD44 adhesion molecule expression in renal cell carcinoma. Hum Pathol. 2001;32: 1209-1215. 9. Schraml P, Struckmann K, Hatz F, et al. VHL mutations and their correlation with tumour cell proliferation, microvessel density, and patient prognosis in clear cell renal cell carcinoma. J Pathol. 2002; 196:186-193. 10. Yao X, Qian CN, Zhang ZF, et al. Two distinct types of blood vessels in clear cell renal cell carcinoma have contrasting prognostic implications. Clin Cancer Res. 2007;13:161-169. 11. Djordjevic G, Mozetic V, Mozetic DV, et al. Prognostic significance of vascular endothelial growth factor expression in clear cell renal cell carcinoma. Pathol Res Pract. 2007;203:99-106. 12. Paradis V, Lagha NB, Zeimoura L, et al. Expression of vascular endothelial growth factor in renal cell carcinomas. Virchows Arch. 2000;436:351-356. 13. Jain RK, Duda DG, Clark JW, et al. Lessons from phase III clinical trials on anti-VEGF therapy for cancer. Nat Clin Pract Oncol. 2006;3:24-40. 14. Escudier B, Eisen T, Stadler WM, et al. Sorafenib in advanced clear-cell renal-cell carcinoma. N Engl J Med. 2007;356:125-134. 15. Llovet J, Ricci S, Mazzaferro V, et al. Sorafenib improves survival in advanced hepatocellular carcinoma (HCC): results of a phase III randomized placebo-controlled trial (SHARP trial). J Clin Oncol. 2007;25; ASCO Annual Meeting Proceedings Part I: LBA1. 16. Motzer RJ, Hutson TE, Tomczak P, et al. Sunitinib versus interferon alfa in metastatic renal-cell carcinoma. N Engl J Med. 2007; 356:115-124. 17. Verheul HM, Pinedo HM. Possible molecular mechanisms involved in the toxicity of angiogenesis inhibition. Nat Rev Cancer. 2007;7:475-485. 18. Rickles FR, Patierno S, Fernandez PM. Tissue factor, thrombin, and cancer. Chest. 2003;124:58S-68S.

613

19. Hofmann R, Lehmer A, Buresch M, et al. Clinical relevance of urokinase plasminogen activator, its receptor, and its inhibitor in patients with renal cell carcinoma. Cancer. 1996;78:487-492. 20. Los M, Zeamari S, Foekens JA, et al. Regulation of the urokinasetype plasminogen activator system by the von Hippel–Lindau tumor suppressor gene. Cancer Res. 1999;59:4440-4445. 21. Wakelin SJ, Marson L, Howie SE, et al. The role of vascular endothelial growth factor in the kidney in health and disease. Nephron Physiol. 2004;98:73-79. 22. Gruber M, Simon MC. Hypoxia-inducible factors, hypoxia, and tumor angiogenesis. Curr Opin Hematol. 2006;13:169-174. 23. Raval RR, Lau KW, Tran MG, et al. Contrasting properties of hypoxia-inducible factor 1 (HIF-1) and HIF-2 in Von Hippel–Lindauassociated renal cell carcinoma. Mol Cell Biol. 2005;25:5675-5686.

614

24. Hughson MD, He Z, Liu S, et al. Expression of HIF-1 and ubiquitin in conventional renal cell carcinoma: relationship to mutations of the von Hippel–Lindau tumor suppressor gene. Cancer Genet Cytogenet. 2003;143:145-153. 25. Kim CM, Vocke C, Torres-Cabala C, et al. Expression of hypoxia inducible factor-1alpha and 2alpha in genetically distinct early renal cortical tumors. J Urol. 2006;175:1908-1914. 26. Na X, Wu G, Ryan CK, et al. Overproduction of vascular endothelial growth factor related to von Hippel–Lindau tumor suppressor gene mutations and hypoxia-inducible factor-1 alpha expression in renal cell carcinomas. J Urol. 2003;170:588-592. 27. Lidgren A, Hedberg Y, Grankvist K, et al. Hypoxia-inducible factor 1alpha expression in renal cell carcinoma analyzed by tissue microarray. Eur Urol. 2006;50:1272-1277.

UROLOGY 75 (3), 2010