Role of urokinase plasminogen activator and its receptor in metastasis and invasion of neuroblastoma

Role of Urokinase Plasminogen Activator and its Receptor in Metastasis and Invasion of Neuroblastoma By Peng Li, Ya Gao, Zongzheng Ji, Xiansheng Zhang, Quan Xu, Gongcai Li, Zhengtuan Guo, Baijun Zheng, and Xinkui Guo Xi’an, China

Background/Purpose: Urokinase plasminogen activator (uPA) is a serine proteinase that has been suggested to play an important role in tumor invasion and metastasis. It binds to a specific membrane receptor, uPA receptor (uPAR), and activates plasminogen to form plasmin, which participates in tissue degradation and proteolysis. Binding of uPA to its receptor accelerates the activation of uPA from pro-uPA, enhancing the activity of the uPA/uPAR cascade. Because of the high metastatic and invasive potential of neuroblastoma (NB) cells, the authors have analyzed in the current study, the concomitant of uPA and its receptor in NB. Methods: The expression and distribution of uPA and uPAR were analyzed by immunostaining in 52 neuroblastoma tissues; at the same time we use the reverse transcriptase polymerase chain reaction (RT-PCR) for neuroendocrine protein gene products 9.5 (PGP 9.5) mRNA to detect small numbers of NB cells in the peripheral blood and bone marrow (BM) and study the relationship uPA and uPAR to the ability of invasion and metastasis of NB cells. To identify risk factors for disease progression, the authors performed a retrospective analysis of clinical (age, sex, and risk group) and tumor biologic markers (histology, MYCN, DNA ploidy, chromosome 1 p, PGP9.5, uPA, uPAR, and combined uPA and uPAR) in all patients. Survival curves were estimated using the Kaplan-Meier method. Univariate analysis was performed with the log-rank test. Multivariate analysis was performed using the Cox proportional hazards regression model. Results: The results of immunohistochemistry showed that uPA and uPAR were localized mainly in the membrane and cytoplasm of tumor cells. The positive rate of uPA in the high-risk group (23 of 25, 92.0%) was remarkably higher than that in intermediate-risk group (8 of 17, 47.1%) and low-risk group (3 of 10, 30.0%), in UH (26 of 29, 89.7%) was higher than in FH (8 of 23, 34.8%), respectively, and statistical significance was remarkable both P ⬍ .01). Similar results were obtained for uPAR. The positive rate of uPAR in the high-risk group (22 of 25, 88.0%) was substantially higher compared

with that in intermediate-risk group (6 of 17, 35.3%) and low-risk group (2 of 10, 20.0%; P ⬍ .01). The positive rate of uPAR in UH (24 of 29, 82.8%) was higher compared with that in FH (6 of 23, 26.1%), and statistical significance was remarkable (P ⬍ .01). PGP9.5 mRNA in peripheral blood and BM was detected in 24 of 52 (45.2%) patients. The positive rate of PGP 9.5 mRNA in peripheral blood and BM in the cases positive for uPA (22 of 34, 64.7%) was markedly higher than that in the cases negative for uPA (11.1%, 2 of 18), and statistical significance was remarkable (P ⬍ .01). There was significant difference in the positive rate of PGP9.5 mRNA between the group positive for uPAR (66.7%, 20 of 30) and the group negative for uPAR (18.2%, 4 of 22), and a larger difference was found between the group positive for both uPA and uPAR (73.1%, 19 of 26) and the group negative for uPA or uPAR (19.2%, 5 of 26). The overall survival (OS) and eventfree survival (EFS) rates at 5 years for all patients were, respectively, 70% ⫾ 3% and 63% ⫾ 3% with a median follow-up of 65 months (range 13 to 20). Among all the biologic and clinical features analyzed, multivariate analysis using Cox proportional hazards regression showed that age, MYCN, and combined uPA and uPAR remained significant predictors for both OS and EFS (P ⬍ .01, respectively). Both EFS rate and OS rate were significantly better for patients who positively expressed uPA and uPAR than those who negatively expressed uPA or uPAR. Conclusions: This study showed that uPA and uPAR were overexpressed in high-risk and UH tumor of NB, and that overexpression of both factors was associated with the ability of invasion, metastasis, and prognosis of NB. The presence of high levels of combined uPA and uPAR may be a new prognostic marker that would allow us to identify patients with poorer prognosis who might benefit from more aggressive surgical and adjuvant treatment. J Pediatr Surg 39:1512-1519. © 2004 Elsevier Inc. All rights reserved. INDEX WORDS: Urokinase plasminogen activator, urokinase plasminogen activator receptor, neuroblastoma.

N From the Department of Pediatric Surgery, the Second Hospital of Xi’an Jiaotong University, Xi’an, China. Address reprint requests to Peng Li, MD, Department of Pediatric Surgery, The Second Hospital, Xian Jiaotong University, 157 West 5 Rd, Shaanxi, Xi’an 710004, PR China. © 2004 Elsevier Inc. All rights reserved. 0022-3468/04/3910-0013$30.00/0 doi:10.1016/j.jpedsurg.2004.06.011 1512

EUROBLASTOMA (NB) is one of the most common pediatric solid tumors originating from the sympathoadrenal lineage of neural crest. This tumor has the propensity to regress spontaneously in some infants or to differentiate into a benign ganglioneuroma in some older patients. Unfortunately, neuroblastoma is metastatic at the time of diagnosis in the majority of patients, and it usually leads to rapid tumor progression and a fatal outcome.

Journal of Pediatric Surgery, Vol 39, No 10 (October), 2004: pp 1512-1519

UROKINASE PLASMINOGEN ACTIVATOR IN NB

The mechanisms that regulate this aggressive growth behavior in neuroblastoma are not very clear. Several studies indicate that during tumor cell invasion and metastasis, proteolytic enzymes may participate in the degradation of extracellular matrix components.1,2 In the past, scientists have focused their attention on the pathways of plasminogen activation. Plasminogen is an inactive proenzyme that can be converted to plasmin by 2 types of plasminogen activators— uPA (urokinase plasminogen activator) and tPA (tissue plasminogen activator). It is tPA rather than uPA that is mainly involved in physiologic activation of plasmin during intravascular thrombolysis.3,4 However, uPA appears to play a pivotal role in pericellular proteolysis during cell migration and tissue remodeling.5 uPA is initially released from various cells as an enzymatically inactive proenzyme (pro-uPA), which can be cleaved by serine proteinases, cysteine proteinases, or thermolysin in its enzymatically active high molecular weight form or by thrombin and granulocyte elastase in enzymatically inactive high molecular weight uPA.6 Both uPA and pro-uPA bind with high affinity to a specific cell-surface receptor (uPA receptor). The uPA receptor (uPAR) is a cysteine-rich glycoprotein with an approximate molecular weight of 55 to 60 kDa. Receptor binding of uPA or pro-uPA strongly accelerates pro-uPA activation and increases the enzymatic activity of uPA itself7 uPA converts the zymogen plasminogen to plasmin, an enzyme that degrades fibrin and a number of other components of the extracellular matrix, such as type IV collagen, fibronectin, and laminin.6 Plasmin also activates latent collagenases to potentiate their lytic activity.6,8 This activation at the cell surface may enable cells to exercise a focal and directional proteolysis of the extracellular matrix. Subsequently, plasmin formation facilitates the passage of migrating cells through tissue barriers.9 Recently, elevated levels of uPA and uPAR have been reported in prostate,10,11 lung,12 ovarian,13 breast14 and gastrointestinal carcinomas.15 In many cases, their increased expression seems to be associated with an increased metastatic potential and poor survival. Because of the high metastatic and invasive potential of neuroblastoma cells, we have analyzed in the current study the concomitant of uPA and its receptor in neuroblastoma. In addition, our previous studies have described the use of the reverse transcriptase polymerase chain reaction (RT-PCR) for neuroendocrine protein gene products 9.5 (PGP 9.5) mRNA to identify NB cells in the peripheral blood and bone marrow (BM) with a higher sensitivity of detection.16 In the current study, we used RTPCR for PGP 9.5 mRNA to detect small numbers of NB cells in the peripheral blood and BM and studied the

1513

relation between uPA system and the ability of invasion and metastasis of NB cells. MATERIALS AND METHODS

Patients and Tissues Samples Neuroblastoma tissues were obtained from 24 girls and 28 boys undergoing surgery or biopsy at the Second Hospital of Xi’an Jiao Tong University (Xi’an, China) between 1993 and 2003. The median age of the NB patients was 4.3 years, with a range of 6 months to 10 years. The median follow-up period after diagnosis for the surviving children was 65 months (range, 13 to 120 months). The neuroblastomas were staged according to the International Neuroblastoma Staging System (INSS).17 Based on the Children’s Oncology Group (COG) Neuroblastoma Risk Stratification System, all patients were classified into low-risk, intermediate-risk and high-risk groups. The tumors were divided further into 2 groups: favorable histopathology (FH) and unfavorable histopathology (UH) on the basis of the International Neuroblastoma Pathology Classification System.18 Freshly removed tumor samples were fixed in paraformaldehyde for 12 to 24 hours and paraffin embedded for histologic analysis. Patients’ blood samples and BM samples were obtained at diagnosis and immediately were frozen in liquid nitrogen and maintained at ⫺80°C until use.

uPA and uPAR Immunohistochemistry Consecutive 3- to 5-␮m paraffin-embedded tissue sections were subjected to immunostaining using the streptavidin-peroxidase technique. Tissue sections were submerged for 15 minutes in Tris-buffered saline (TBS; 10 mmol/L Tris-HCL 0.85% sodium chloride, pH 7.4) containing 0.1% (vol/vol) Triton X-100 and then were washed for 5 minutes in TBS, as previously reported.19,20 Endogenous peroxidase activity was blocked by incubating the slides in methanol and in methanol-0.6% hydrogen peroxide followed by 3 washings in methanol and TBS containing 0.1% bovine serum albumin (BSA).19 After treatment with hyaluronidase (1 mg/mL- in 100 mmol/L sodium acetate, 0.85% sodium chloride), the sections were incubated for 30 minutes at 37°C with 10% normal goat serum before overnight incubation at 4°C with the specific monoclonal antibodies (Huamei Biomedicals, China) diluted in 10% normal goat serum. Bound antibody was detected with a biotinylated goat antimouse IgG secondary antibody and a streptavidinperoxidase complex. This was followed by incubation with diaminobenzidene tetrahydrochloride (0.05%) as the substrate and then counterstaining with Mayer’s hematoxylin. To ensure specificity of the immunostaining reactions, consecutive sections were incubated either in the absence of the primary antibody or with a nonimmunized mouse IgG antibody. In both cases, no immunostaining was detected. Histopathologic analysis of the immunohistochemical results was performed by 2 independent pathologists blinded to patient status, followed by resolution of any differences by joint review and consultation with a third

1514

PENG ET AL

observer. The intensity of the immunohistochemical signal (intensity score) was stratified into 2 groups: negative (no or weak immunostaining) and positive (moderate or intense immunostaining).

D1Z2 or PCR-based microsatellites.24 Tumor cell DNA content was analyzed using single-cell image cytometry or flow cytometry.25 Near-triploid DNA content was considered if the DNA index ranged from 1.25 to 1.75.

RT-PCR for PGP 9.5

Statistical Analysis

Peripheral blood mononuclear cells (PBMNC) and bone marrow mononuclear cells (BMMNC) were isolated by density centrifugation. Total cellular RNA was extracted from patients’ PBMNC and BMNNC by a modification of acid guanidinium thiocyanate-phenolchloroform (AGPC). RNA was washed in ethanol and dissolved in diethyl pyrocarbonate (DEPC)-treated water. Recovered RNA and its purity were assessed by optical densimetry at 260 and 280 nm. The primers for PGP 9.5, the sequences of which were 5⬘-AGA TCA ACC CCG AGA TGC ACA AAG (forward) and 3⬘-ATT AGG CTG CCT TGC AGA GAG CCA CGG CAG A, were synthesized by Rikaken Co, Japan. RNA LA PCR kit (AMV; Code No. PRO12A, TaKaRa Biomedicals, Japan) was used to perform RTPCR. The kit included random primer, internal control primer, and positive primer. The RT reaction mixture consisted of 5 ␮L MgCl2, 2 ␮L RNA-PCR buffer (10⫻), 2.5 ␮L DEPC H2O, 8 ␮L dNTP mixture (dGTP, dCTP, dTTP, and dATP), 0.5 ␮L RNA guard, 1 ␮L random primers and 1 ␮L reverse transcriptase to a total volume of 20 ␮L. Then, 2 ␮L RNA of each sample (routinely 2 ␮g) was heated to 55°C for 4 minutes before cooling and adding the above mix. After mixing, the reaction was allowed to proceed at 42°C for 1 hour. The PCR mixture consisted of 6 ␮L MgCl2, 8 ␮L LA PCR buffer II (10⫻), 63.5 ␮L DEPC dH2O, 0.5 ␮L Taq polymerase (TaKaRa Biomedicals, Japan), and 1 ␮L each pair of primers (100 pmol) to a total volume of 80 ␮L. The PCR conditions were 40 seconds at 94°C and 75 seconds at 72°C for 35 cycles, followed by a 10-minute extension at 72°C RT-PCR product (10 ␮L) was electrophoresed on a 1% agarose gel and visualized by ethidium bromide staining. For each test, both negative and positive controls were included. Molecular weight (MW) Marker 4 (␾ ⫻ 174DNA/HaeIII) were purchased from Wakan Pharmaceutics Co., Japan. The positive band lied in 653 base pairs.

To compare variables of interest, the ␹2 test was used where appropriate. Event-free survival (EFS) and overall survival (OS) estimates were determined using the Kaplan-Meier method, and survival curves were compared using the log-rank test. Multivariate analysis by means of a Cox proportional hazards regression model was used to assess the relation between survival and covariates. Differences were considered significant if P was less than .05. All statistical analyses were 2 tailed and were performed using Statistical Analysis System software (SAS Institute, Cary, NC).

Genetic Studies MYCN copy number was analyzed using FISH21 or Southern blotting22 in all tumors and considered amplified if the copy number per haploid genome was 3 or more. All tumors were investigated for chromosome 1 integrity with fluorescence in situ hybridization (FISH) as described previously23 using probe D1Z1 for chromosome 1 centromeres and the 1p36.33-specific probe

RESULTS

Table 1 summarized the clinical and biologic characteristics of all 52 patients in which 10 were the low risk, 17 were the intermediate risk, and 25 were the high risk. Eighteen patients were younger than 1 year of age at diagnosis. Twenty-nine cases for which pathology classification was applicable had unfavorable histology, and 23 cases had favorable histology. DNA index was in the diploid-tetraploid range in 22 patients and hyperdiploid (near-triploid) in 30 patients. MYCN copy number was amplified in 12 cases and nonamplified (normal) in 40 cases. Normal and deleted chromosome 1p were found in 40 and 12 cases, respectively. The results of immunohistochemistry showed that uPA and uPAR were localized mainly in the membrane and cytoplasm of tumor cells (Figs 1 & 2). Of 52 patients, 34 showed positive uPA results (65.4%), and 30 showed positive uPAR results (57.7%), both uPA- and uPAR-positive results were found in 26 cases (50.0%). Table 2 showed that the positive rate of uPA in the high-risk group (23 of 25, 92.0%) was remarkably higher than that in intermediate-risk group (8 of 17, 47.1%) and the low-risk group (3 of 10, 30.0%), in UH (26 of 29, 89.7%) was higher than in FH (8 of 23, 34.8%), respectively, and statistical significance was remarkable; both P ⬍ .01. Similar results were obtained for uPAR; the positive rate of uPAR in the high-risk group (22 of 25, 88.0%) was substantially higher compared with that in intermediate-risk group (6 of 17, 35.3%) and the low-risk group (2 of 10, 20.0%); P ⬍ .01. The positive rate of uPAR in UH (24 of 29, 82.8%) was higher compared with that in FH (6 of 23, 26.1%), and statistical significance was remarkable; P ⬍ .01. Table 3 showed that PGP9.5 mRNA in peripheral blood and BM was detected in 24 of 52 (45.2%) patients.

UROKINASE PLASMINOGEN ACTIVATOR IN NB

1515

Table 1. Statistical Univariate and Multivariate Analysis of Biological and Clinical Features Investigated on OS and EFS Total (n ⫽ 52)

Sex Male Female Age at diagnosis ⬍12 mo ⬎12 mo MYCN Normal Amplified Chromosome 1 p Normal Deleted DNA ploidy 3n 2n/4n Risk group Low and intermediate risk High risk Histopathology FH UH uPA uPA (⫹) uPA (⫺) uPAR uPAR (⫹) uPAR (⫺) Combined uPA and uPAR Both uPA (⫹) and uPAR (⫹) uPA (⫺) or uPAR (⫺) PGP9.5 PGP9.5 (⫹) PGP9.5 (⫺)

P Value for OS

P Value for EFS

Univariate

Multivariate

Univariate

Multivariate

28 24

NS

NS

NS

NS

18 34

⬍.01

⬍.01

⬍.01

⬍.01

40 12

⬍.01

⬍.01

⬍.01

⬍.01

35 17

NS

NS

⬍.05

NS

30 22

⬍.05

NS

⬍.01

NS

27 25

⬍.01

NS

⬍.01

NS

23 29

⬍.01

NS

⬍.01

NS

34 18

⬍.01

NS

⬍.01

NS

30 22

⬍.01

NS

⬍.01

NS

26 26

⬍.01

⬍.01

⬍.01

⬍.01

24 28

NS

NS

⬍.01

NS

NOTE. P values based on log Rank test. Abbreviation: NS, not significant.

Fig 1. Results of immunohistochemistry show that uPA was localized mainly in the membrane and cytoplasm of tumor cells (Streptavidin-peroxidase technique, original magnification ⴛ400.)

Fig 2. Results of immunohistochemistry show that uPAR was localized mainly in the membrane and cytoplasm of tumor cells. (Streptavidin-peroxidase technique, original magnification ⴛ400.)

1516

PENG ET AL

Table 2. Relations of uPA and uPAR Positive Rate to Tumor Risk Group and Type of Histopathology

Risk group low-risk Intermediate-risk high-risk Histopathology FH UH

No. of Patients

uPA(⫹)

P Value

uPAR(⫹)

P Value

10 17 25

3 (30.0%) 8 (47.1%) 23 (92.0%)

⬍.01

2 (20.0%) 6 (35.3%) 22 (88.0%)

⬍.01

23 29

8 (34.8%) 26 (89.7%)

The positive rate of PGP 9.5 mRNA in peripheral blood and BM in the cases positive expressed for uPA (22 of 34, 64.7%) was markedly higher than that in the cases negatively expressed for uPA (11.1%, 2 of 18), and statistical significance was remarkable; P ⬍ .01. There was significant difference in the positive rate of PGP9.5 mRNA between the group positively expressed for uPAR (66.7%, 20 of 30) and the group negatively expressed for uPAR (18.2%, 4 of 22), and a larger difference was found between the group both positively expressed for uPA and uPAR (73.1%, 19 of 26) and the group negatively expressed uPA or uPAR (19.2%, 5 of 26). The overall survival (OS) and event-free survival (EFS) rates at 5 years for all patients were 70% ⫾ 3% and 63% ⫾ 3%, respectively. Table 1 summarized the relevant results of univariate analysis. Among all the biologic and clinical features analyzed, age, MYCN, DNA ploidy, risk group, histology, uPA, uPAR, and combined uPA and uPAR were significant prognostic variables when OS was analyzed (log-rank test, P ⬍ .05), but gender, PGP9.5, and chromosome 1p were not significant. Age, MYCN, DNA ploidy, risk group, histology, uPA, uPAR, combined uPA and uPAR, PGP9.5, and chromosome 1p were significant prognostic variables when EFS was analyzed (log-rank test, P ⬍ .05), and only gender was not significant. Multivariate analysis using Cox proportional hazards regression showed that age, MYCN, and combined uPA and uPAR remained to be significant predictors for both OS and EFS (P ⬍ .01). Both EFS rate and OS rate were significantly better for patients who both positively expressed uPA and uPAR than those negative expressed uPA or uPAR (Figs 3 & 4).

⬍.01

6 (26.1%) 24 (82.8%)

⬍.01

DISCUSSION

An important clinical characteristic of NB is early metastasis to lymph nodes and distant organs, and the development of distant metastases is decisive for the fate of patients with NB. However, the mechanisms that contribute to the ability of NB cells during tumor cell invasion and metastasis are not well understood. The use of RT-PCR for the detection of small numbers of circulating tumor cells has been described in other literature.26,27 Expression of PGP 9.5 mRNA has been used to detect NB cells in BM and peripheral blood with a reported sensitivity of 1 cell per 107 peripheral blood mononuclear cells.28 Our previous literature16 reported that RT-PCR for PGP 9.5 mRNA may be a useful, noninvasive method for the early detection of disease dissemination and monitoring of disease status. Positive PGP9.5 mRNA showed that NB cells were present. In this study, we used PGP9.5 mRNA as a marker for whether NB cells had metastasized and then studied the relation between the uPA systems and the ability of NB cells invade metastasize. Activation of the uPA plasminogen/plasmin system seems to contribute to tumor aggressiveness and shorter postoperative survival. In the current study, we examined the relation between the uPA system and the tumor stage, type of histopathology of NB, and PGP 9.5 mRNA expressed by NB cells in the peripheral blood and BM. Then we performed a retrospective analysis to identify clinical and biologic features that may be associated with adverse clinical outcome for these patients. This study found that uPA and uPAR were localized mainly in the membrane and cytoplasm of tumor cells. This observation was most likely relevant to human cell

Table 3. Relation of uPA and uPAR Positive Rate and PGP 9.5 mRNA in the Peripheral Blood and BM

uPA(⫹) uPA(⫺) uPAR(⫹) uPAR(⫺) uPA(⫹) and uPAR(⫹) uPA(⫺) or uPAR(⫺)

No. of Patients

PGP9.5 (⫹)

PGP9.5 (⫺)

Positive Rate (%)

P Value

34 18 30 22 26 26

22 2 20 4 19 5

12 16 10 18 7 21

64.7 11.1 66.7 18.2 73.1 19.2

⬍.01 ⬍.01 ⬍.01

UROKINASE PLASMINOGEN ACTIVATOR IN NB

Fig 3. Kaplan-Meier OS curve for patients both positively expressed for uPA and uPAR and patients negatively expressed for uPA or uPAR.

progression, because the binding of pro-uPA to uPAR initiated uPA-dependent proteolytic activity on the surface of invading cells at the invasive foci. Increased stromal release of uPA may result in increased binding to uPAR in tumor cells as well as capillary endothelial cells. Furthermore, ligand binding to uPAR increased its affinity to other cell surface proteins, eg, vitronectin and integrin receptors, and initiated assembly of these proteins together with caveolin and src kinases at focal adhesion sites.29-31 Focal adhesion sites mediated connection between the cytoskeleton and the extracellular matrix in the leading edge of the cell during cell migration. Also, intracellular signaling during cell migration was initiated by integrin receptors and involved phosphorylation of src kinases, focal adhesion kinase, and, subsequently, the mitogen-activated protein kinase pathway.32 Both migration and pericellular proteolysis was initiated by ligand binding to uPAR and was expressed by invasive cell phenotypes like tumor cells and endothelial cells during neo-angiogenesis. Thus, the stoma release of uPA may facilitate and stimulate tumor cell invasion as well as concomitant angiogenesis.33 The current study showed that there was a substantially higher expression of uPA and uPAR in high-risk tumors compared with that in intermediate-risk and lowrisk tumors and in UH compared with in FH. Furthermore, there was a significant difference in the positive rate of PGP9.5 mRNA in peripheral blood and BM between the cases positively expressed for uPA and the cases negatively expressed for uPA. Similar results were obtained between the cases positively expressed for uPAR and the cases negatively expressed for uPAR. A bigger difference was found between the group both positively expressed for uPA and uPAR and the group negatively expressed uPA or uPAR. Clinical variable of age that was recognized to be a strong prognostic feature for NB in general34 also showed significance in our study. We found that infants less than 1 year of age have significantly better OS and EFS rate than older children.

1517

MYCN amplification occurred in approximately 20% of primary NB tumors and was strongly associated with the presence of metastatic disease and poor prognosis.35,36 These observations suggested that MYCN critically contributed to the clinically aggressive behavior of high-risk NB tumors, and laboratory studies supported this hypothesis.37,38 The level of expression of MYCN has been shown to directly correlate with the growth potential of NB cells in vitro as well as in vivo.39-40 A role for MYCN in NB pathogenesis was supported further by our studies showing that MYCN amplification remained to be a significant predictor for both OS and EFS by using Cox proportional hazards regression multivariate analysis. Although previous articles have reported that DNA ploidy or chromosome 1p has been validated as an independent prognostic variable in NB,41-45 in our study, their significance to OS and EFS rate was lost when combined uPA and uPAR were included in the regression model. In our series, combined uPA and uPAR were powerful markers for OS and EFS. Both EFS rate and OS rate were significantly better for patients who both positively expressed uPA and uPAR than those who negatively expressed uPA or uPAR (Figs 3 & 4). The presence of both the ligand and the receptor in the same tumor cells indicates that NB cells may have the ability to upregulate plasminogen activation by uPA/ uPAR production. Binding of uPA to its receptor enhances the enzymatic activity of uPA, which then accelerates the activation of plasminogen to plasmin. In addition to these mechanisms. uPAR-bound uPA is involved in the activation of growth factors. Besides hepatocyte growth factor (HGF) activation which can be catalyzed by uPA directly, activation of transforming growth factor betas (TGF-␤s) from their proforms, basic fibroblast growth factor (bFGF) mobilization, and extracellular matrix degradation are mediated by plasmin and

Fig 4. Kaplan-Meier EFS curve for patients both positively expressed for uPA and uPAR and patients negatively expressed for uPA or uPAR.

1518

PENG ET AL

by uPAR-bound or free uPA.46 These multifactorial effects of uPA/uPAR on growth factors and the increase in proteolysis might enhance the ability of NB cells to migrate, invade mormal tissue, and metastasize. These effects may be increased by the additional expression of uPA and uPAR in the surrounding stroma adjacent to tumor cells. But combined uPA and uPAR were not the only prognostic factors that showed that the mechanism of metastasis and invasion of tumor cells were complicated. No prognostic variables can explain it thoroughly. The

relationship between the uPA system and other prognostic variables needs to be studied further. This study showed that uPA and uPAR were overexpressed in high-risk and UH tumor of NB and that overexpression of both factors was associated with the ability of invasion, metastasis, and a poor prognosis for NB. The presence of high levels of combined uPA and uPAR may be a new prognostic marker that would allow us to identify patients with a poorer prognosis who might benefit from more aggressive surgical and adjuvant treatment.

REFERENCES 1. Fidler IJ, Gersten DM, Hart IR: The biology of cancer invasion and metastasis. Adv Cancer Res 28:149-150, 1978 2. Chikahisa L, Matsuo K, Yamada Y: Modulation of tumor cell motility by plasmin. Invasion Metastasis 17:323-333, 1997 3. Lyden PD, Zivin JA, Clark WA, et al: Tissue plasminogen activator-mediated thrombolysis of cerebral emboli and its effect on hemorrhagic infarction in rabbits. Neurology 39:703-708, 1989 4. Gupta AA, Leaker M, Andrew M, et al: Safety and outcomes of thrombolysis with tissue plasminogen activator for treatment of intravascular thrombosis in children. J Pediatr 139:682-688, 2001 5. Rettenberger P, Wilhelm O, Oi H, et al: A competitive chromogenic assay to study the functional interaction of urokinase-type plasminogen activator with its receptor. Biol Chem Hoppe Seyler 376:587594, 1995 6. Schmitt M, Janicke F, Moniwa N, et al: Tumor-associated urokinase-type plasminogen activator: Biological and clinical significance. Biol Chem Hoppe Seyler 373:611-622, 1992 7. Plesner T, Behrendt N, Ploug M: Structure, function and expression on blood and bone marrow cells of the urokinase-type plasminogen activator receptor, uPAR. Stem Cells 15:398-408, 1997 8. Vassalli J-D, Sappino AP, Belin D: The plasminogen activator’ plasmin system. J Clin Invest 88:1067-1072, 1991 9. Xue W, Mizukami I, Todd RF 3rd, et al: Urokinase-type plasminogen activator receptors associate with beta1 and beta3 integrins of fibrosarcoma cells: Dependence on extracellular matrix components. Cancer Res 57:1682-1689, 1997 10. Festuccia C, Dolo V, Guerra F, et al: Plasminogen activator system modulates invasive capacity and proliferation in prostatic tumor cells. Clin Exp Metastasis 16:513-528, 1998 11. Miyake H, Hara I, Yamanaka K, et al: Elevation of serum levels of urokinase-type plasminogen activator and its receptor is associated with disease progression and prognosis in patients with prostate cancer. Prostate 39:123-129, 1999 12. Volm M, Mattern J, Koomagi R: Relationship of urokinase and urokinase receptor in non-small cell lung cancer to proliferation, angiogenesis, metastasis and patient survival. Oncol Rep 6:611-615, 1999 13. Tecimer C, Doering DL, Goldsmith LJ, et al: Clinical relevance of urokinase-type plasminogen activator, its receptor and inhibitor type 1 in ovarian cancer. Int J Gynecol Cancer 10:372-381, 2000 14. Borstnar S, Vrhovec I, Svetic B, et al: Prognostic value of the urokinase-type plasminogen activator, and its inhibitors and receptor in breast cancer patients. Clin Breast Cancer 3:138-146, 2002 15. Okusa Y, Ichikura T, Mochizuki H, et al: Urokinase type plasminogen activator and its receptor regulate the invasive potential of gastric cancer cell lines. Int J Oncol 17:1001-1005, 2000 16. Gao Ya, Li Gongcai, Zhang Xiansheng, et al: Detection of neuroblastoma cells in blood by reverse transcriptase ployerase chain reaction. Chin Med J 110:341-345, 1997

17. Brodeur GM, Pritchard J, Berthold, et al: Revisions of the international criteria for neuroblastoma diagnosis, staging, and response to treatment. J Clin Oncol 11:1466-1477, 1993 18. Shimada H, Ambros IM, Dehner LP, et al: The international Neuroblastoma Pathology Classification (the Shimada system). Cancer 86:364-372, 1999 19. Friess H, Yamanaka Y, Buchler M, et al: Enhanced expression of transforming growth factor beta isoforms in pancreatic cancer correlates with decreased survival. Gastroenterology 105:1846-1856, 1993 20. Miturski R, Semczuk A, Tomaszewski J, et al: bcl-2 protein expression in endometrial carcinoma: The lack of correlation with p53. Cancer Lett 13(133):63-69, 1998 21. Ambros IM, Zellner A, Roald B, et al: Role of ploidy, chromosome 1p, and Schwann cells in the maturation of neuroblastoma. N Engl J Med 334:1505-1511, 1996 22. Seeger RC, Brodeur GM, Sather H, et al: Association of multiple copies of the N-myc oncocene with rapid progression of neuroblastoma. N Engl J Med 313:1111-1116, 1985 23. Ambros PF, Ambros IM, Strehl S, et al: Regression and progression in neuroblastoma. Does genetics predict tumor behavior? Eur J Cancer 31A:510-515, 1995 24. Martinsson T, Sjoberg RM, Hedborg F, et al: Deletion of chromosome 1 p loci and microsatellite instability in neuroblastomas analyzed with short-tandem repeat polymorphisms. Cancer Res 55: 5681-5686, 1995 25. Falkmer UG, Falkmer S: The value of cytometric DNA analysis as a prognostic tool in neuroendocrine neoplastic diseases. Pathol Res Pract 191:281-303, 1995 26. Takeda S, Nakao A: Genetic detection and clinical applications in patients with hepatocellular carcinoma. Nippon Geka Gakkai Zasshi 103:472-475, 2002 27. Shin JH, Chung J, Kim HO, et al: Detection of cancer cells in peripheral blood of stomach cancer patients using RT-PCR amplification of tumour-specific mRNAs. Aliment Pharmacol Ther 16(suppl 2):137-144, 2002 28. Mattano JA, Moss TJ, Emerson SG, et al: Sensitive detection of circulating neuroblastoma cells by reverse transcriptase-polymerase chain reaction. Cancer Res 52:4701, 1992 29. Andreasen PA, Egelund R, Petersen HH: The plasminogen activation system in tumor growth invasion and metastasis. Cell Mol Life Sci 57:25-40, 2000 30. Wei Y, Yang X, Simon DI, et al: A role for caveolin and the urokinase receptor in integrin-mediated adhesion and signaling. J Cell Biol 144:1285-1294, 1999 31. Chambers SK, Ivins CM, Carcangiu ML: Plasminogen activator inhibitor-1 is an independent poor prognostic factor for survival in advanced stage epithelial ovarian cancer patients. Int J Caner 79:449454, 1998

UROKINASE PLASMINOGEN ACTIVATOR IN NB

32. Danen EH, Lafrenie RM, Miyamoto S, et al: Integrin signaling: Cytoskeletal complexes, MAP kinase activation and regulation of gene expression. Cell Adhes Commun 6:217-224, 1998 33. Berger DH: Plasmin/Plasminogen system in colorectal cancer. World J Surg 26:767-771, 2002 34. Maris JM, Matthay KK: Molecular biology of neuroblastoma. J Clin Oncol 17:2264-2279, 1999 35. Brodeur GM, Seeger RC, Sather H, et al: Amplification of N-myc in untreated human neuroblastoma correlates with advanced disease stage. Science 224:1121-1124, 1984 36. Seeger RC, Brodeur GM, Sather H, et al: Association of multiple copies of the N-myc oncogene with rapid progression of neuroblastoma. N Engl J Med 313:1111-1116, 1985 37. Schweigerer L, Breit S, Wenzel A, et al: Augmented MYCN expression advances the malignant phenotype of human neuroblastoma cells: Evidence for induction of autocrine growth factor activity. Cancer Res 50:4411-4416, 1990 38. Gross N, Miescher G, Beck K, et al: Altered growth and phenotype in clonal mycN transfectans of the SK-N-SH neuroblastoma cell line. Int J Cancer 59:141-148, 1994 39. Schmidt ML, Salwen HR, Manohar CF, et al: The biological effects of antisense N-myc expression in human neuroblastoma. Cell Growth Differ 5:171-178, 1994 40. Negroni A, Scarpa S, Romeo A, et al: Decrease of proliferation

1519

rate and induction of differentiation by a MYCN antisense DNA oligomer in a human neuroblastoma cell line. Cell Growth Differ 2:511-518, 1991 41. Look AT, Hayes FA, Nitschke R, et al: Cellular DNA content as a predictor of response to chemotherapy in infants with unresectable neuroblastoma. N Engl J Med 311:231-235, 1984 42. Taylor SR, Locker J: A comparative analysis of nuclear DNA content and N-myc gene amplification in neuroblastoma. Cancer 65: 1360-1366, 1990 43. Bourhis J, Devathaire F, Wilson GD, et al: Combined analysis of DNA ploidy index and N-myc genomic content in neuroblastoma. Cancer Res 51:33-36, 1991 44. Gansler T, Chatten J, Varello M, et al: Flow cytometric DNA analysis of neuroblastoma. Correlation with histology and clinical outcome. Cancer 58:2453-2458, 1986 45. Bown N, Cotterill S, Lastowska M, et al: Gain of chromosome arm 17p and adverse outcome in patients with neuroblastoma. N Engl J Med 340:1954-1961, 1999 46. Mori T, Abe T, Wakabayashi Y, et al: Up-regulation of urokinase-type plasminogen activator and its receptor correlates with enhanced invasion activity of human glioma cells mediated by transforming growth factor-alpha or basic fibroblast growth factor. J Neurooncol 46:115-123, 2000