Introduction

Bladder carcinoma is the commonest malignant tumor in Egypt. It constitutes 30.3% of all cancers, 40.6% of male cancers and 14.3% of female cancers. The association of bladder cancer and schistosomiasis has long been evident in many epidemiologic studies (1). Identification of patients with bladder carcinoma at high risk of tumor invasion and early metastasis is mandatory to provide an optimal specific dealing with those patients. Prognostic markers may be the tools to accomplish this. Hopefully advanced methods and new clinically valuable prognostic probes could be evolved from better understanding the molecular basis of tumor progress and metastasis.

Urokinase plasminogen activator (uPA) that is an intracellular serine protease, playing an important role in degradation of extracellular matrix and modulation of cell adhesion and migration leading to local invasion and seeding of tumor cells (2). Urokinase-type plasminogen activator (uPA) is converting cell-bound plasminogen to plasmin and generating the pericellular proteolytic activity. Over expression of uPA has been reported for many carcinoma cell lines and malignant tumors, including: soft tissue sarcoma, osteosarcoma, astrocytoma, melanoma, lung, breast, ovarian and bladder carcinomas (3).

The present study evaluated the prognostic significance of uPA as tumor marker in primary bladder cancers and correlating its expression with other established clinicopathological prognostic parameters and estimating its effect on patients' survival.

Materials and methods

Tissue specimens from 48 Egyptian patients taken during the period between 2000-2001 have bladder carcinoma as well as 6 cystitis cases as a comparative group were examined. Sections had been prepared from formalin- fixed, paraffin- embedded blocks, to study histopathological features by routine haematoxyline & eosin (H&E) examination and to detect mRNA-uPA content by in situ hybridization technique.

Histopathologic Examination

Results

The fifty four cases studied were divided into: -

Malignant group (48): formed of 19 (39.6%) cases of transitional cell carcinoma, 19 (39.6%) cases of squamous cell carcinoma and 6 (12.5%) cases of adenocarcinoma and 4 (8.3%) undifferentiated carcinoma.

Cystitis group (6): as a comparative group.

Histopathological results

Immunohistochemical results (Figures 1-4)

Regarding uPA expression, it was found in (30/48) 75% of the malignant group but in (1/6) 16.7% of cystitis group and the difference was statistically significant (P>0.01).

Distribution of uPA protein expression in the malignant group (Focal, diffuse) showed that 9/36 (25%) were focally distributed and 2/36 (5.5%) were diffusely distributed and there was no statistically significant difference among the studied malignant subgroups while in cystitis groups uPA expression was very weak focal in the positive case.

However, It was observed that high grade (P=0.001), positive lymph node involvement (P=0.027) and high tumor stage (P=0.003) were statistically associated with uPA mRNA expression. While other parameters showed no statistical association with uPA expression table (3).

In cases of TCC; positive vascular invasion (P=0.02), high grade (P=0.01) and high stage (P=0.03) showed statistically significant association with increased mRNA uPA expression while in SCC cases; only the high grades (P=0.02) and the high stages (P=0.04) of the tumor showed this statistical association Table (4).

By using Kaplan Meier equation the factors responsible for the overall survival were the following: Tumor expression of uPA (P=0.03), Stage of tumor (P=0.003), Grade of tumor (P= 0.01), State of lymph node involvement (P=0.02), Vascular invasion (P=0.03) Table (5). By using Cox regression test to asses the most independent factor on patients' survival; the stage of tumor was the most significant parameter. In TCC; tumor grade (P=0.03), Stage (P=0.01) and positive uPA expression (P=0.02) were responsible for overall survival. Only the lymph node status was the determinant factor of overall survival in SCC subgroup (P=0.04) (Chart 1).

Figure 10

Chart 1: Kaplan Meier Survival Analysis for tumor expression of uPA showed significant association between its expression and patients' survival

Figure 2

Table (2): Pathological data of bilharzial and non-bilharzial cases in the studied malignant groups

Figure 3

Table (3): uPA expression and pathological data in the malignant group studied

Figure 4

Table (4): uPA expression and pathological data in TCC and SCC subgroups studied

Figure 5

Table (5): Survival analysis in the studied malignant group

Figure 6

Figure 1: A case of squamous cell carcinoma, grade II showing diffuse nucleocytoplasmic expression of uPA protein that more expressed at the invading margin of tumor. (In situ hybridization, BCIP, eosin counterstain X400).

Figure 7

Figure 2: A case of adenocarcinoma, grade II showing nucleocytoplasmic expression of uPA protein. (In situ hybridization, BCIP, eosin counterstain X400).

Figure 8

Figure 3: A case of undifferentiated carcinoma showing uPA protein expression. (In situ hybridization, BCIP, eosin counterstain X400).

Figure 9

Figure 4: A case of SCC GII showing diffuse cytoplasmic expression of uPA protein. (In situ hybridization, BCIP, eosin counterstain X200).

Survival analysis according to tumor expression of uPA

Discussion

Histopathologic examination of the 48 bladder carcinoma lesions revealed 19 (39.5%) transitional cell carcinoma, 19 (39.5%) squamous cell carcinoma, 6 (12.5 %) adenocarcinoma and 4 (8.5%) undifferentiated carcinoma. El bolkainy in 1998 reported that TCC contributed about 22% of bladder carcinoma and SCC contributed about 59% (10) while Mokhtar et al in 2007 demonstrated that TCC representd 64.2% while SCC constituted 28.3% of their cases (11). These could indicate an increased incidence of TCC cases in Egypt in the last few years. This might be attributed to a real increase of TCC or to the careful recognition of TCC with areas of squamous metaplasia that previously diagnosed as SCC (12).

In the studied specimen from the Egyptian patients, in this study most of the cases were detected in high grade and advanced stage where 29/48 (60.4%) of the malignant cases were grade II and 13/48 (27.1%) were grade III. Moreover, 24/48 (50%) of the malignant cases studied represented stage P3b and 5/48 (10.4%) stage P4a. This clarifies the delay in diagnosis and probably predominance of bilharziasis, that mask the diagnosis in most of the cases (4).

In studied cases, no significant differences have been existed between the different histologic subtypes (TCC, SCC, adenocarcinoma and undifferentiated carcinoma) and age & sex of the patients gross picture, apoptotic index, perineural invasion, lymph node status and associated bilharziasis. However there were significant differences among the studied malignant subgroups as regarding: grade of tumor (P=0.006), mitotic count (P=0.04), vascular invasion (P=0.03) and tumor stage (P=0.01).

Associated schistosomiasis was detected at 2/6 (33.3%) of studied cystitis cases and in 27/48 (56.3%) of bladder carcinoma cases studied. The causal association between bilharziasis and bladder carcinoma was previously explained in many studies. Moreover, several studies from Egypt and other sites of the world had incriminated bilharziasis as a cause of bladder carcinoma (10,11). Therefore Egyptian bladder carcinoma can be classified into two main groups; namely the shistosomiasis and non- shistosomiasis categories. In the present study, the clinicopathological parameter was correlated with the bilharzial and non-bilharzial groups and it was found that the presence of bilharziasis augment the tumor stage in contrast to the other parameters that showed no statistical significance between both groups. This corresponds to the effect of bilharziasis described by other study(13)

The decline in the prevalence of bilharziasis in Egypt during the past decade was associated with significant decrease in frequency of SCC and increase in TCC. Moreover, Egypt starts to be one of the developing industrial countries. Industrialization is one of the most important risk factors in developing bladder carcinoma especially TCC (1).

In this study, 36/48 (75%) of the malignant lesions expressed the uPA gene in contrast to 1/6 (16.7%) of cystitis that expressed the gene in a very weak focal intensity form (P=0.01). This means that levels of uPA in malignant tumors are higher than in the corresponding non-tumourous tissue and goes with previous reports of (2) on their studies on TCC. They discovered that 95% of TCC samples showed higher uPA expression by ISH technique. Also, (1), detected thirty six out of forty six (78.2%) cases of bladder carcinoma expressed the uPA protein by immunohistochemical study. However, in the comparative studied group (1/6) only one case out of six cases showed weak positivity for uPA expression and this came in contrast to the study of Li and his colleagues where they demonstrated that the normal bladder tissue did not express uPA by immunohistochemical method9140. This could be explained by the fact that ISH is a more sensitive and specific technique. Moreover, this studied positive case was cystitis with sevyere dysplasia that could progress to carcinoma later on. Previous study (15) have reported that cases of severe dysplasia with positive uPA cytoplasmic expression in wash out cytology were associated with advanced stages of bladder carcinoma after radical cystectomy procedure.

In our study, the distribution of uPA expression either focal or diffuse showed no statistical difference in the malignant subgroups (P>0.05). The diffuse expression was detected in 27cases out of 36 positive cases (75%) and focal expression was 25%. This means 3/4 of malignant cases acquired the uPA gene to gain the capacity to invade the surrounding tissues, as previously mentioned by Mc Garvey et al (16) when they concluded that the uPA gene is a component of the invasive phenotype of bladder carcinoma.

More interestingly, the present study found that the strong expression of uPA was significantly associated with several factors related to tumor metastasis and the resulting poorer prognosis such as advanced tumor grade (P=0.001), the presence of positive lymph nodes (P=0.02) and advanced tumor stage (P=0.003). However, the protein expression showed no association with age, sex, gross picture, mitotic or apoptotic count, perineural invasion and associated bilharziasis.

uPA positivity represent 36/48 (75%) in overall malignant cases, 15/19 (78.9%) in TCC, 13/19 (68.4%) in SCC, 4/6 (66.6%) in adenocarcinoma and 4/4 (100%) in undifferentiated carcinoma and this goes with previous studies by (2), where they found (78.2%) in their studied on TCC cases and (1), (95%) in their TCC cases studies and in 73% on their studies on SCC.

In TCC studied subgroup, strong uPA protein expression is significantly correlated with tumor grade (P=0.01), vascular invasion (P=0.02) and tumor staging (P=0.03). While in SCC studied subgroup, increased uPA expression is associated with higher tumor grade (P=0.02) and advanced tumor stage (P=0.04). From the previous data we can conclude that the increased tumor grade and advanced tumor stage were the fixed correlating parameters with uPA expression. Therefore, uPA expression could be considered as one of the “hallmarks” of bladder cancer and appear to play roles of varying significance in tumor progression (17).

Considering the available survival analysis in the malignant group, it showed that tumor type (P=0.03), tumor grade (P=0.01), tumor stage (P=0.003), lymph node status (P=0.02), vascular invasion (P=0.03) and positive uPA expression (P=0.03) were independent predictors for cancer specific survival. In patients with undifferentiated carcinoma, high tumor grade, advanced tumor stage, positive lymph node involvement, evidence of vascular invasion and positive uPA protein expression have worse overall outcome than other patients. These results goes with the data previously mentioned by Frazier his colleagues (18), who reported that tumor histopathological type, grade, the presence of positive lymph node, positive surgical margins were the independent factors in patients survival. Using Cox-regression multivariate analysis to detect the most independent factor in the prognosis of bladder carcinoma, the present study showed that the stage was the most independent parameter in patients' outcome. The same result was reported by Kruth et al (19), who noticed that the anatomic extent of tumor namely, the stage was universally accepted as the most important prognostic factor.

Considering the available survival analysis in TCC (19 cases) subgroup, it showed that grade (P=0.03), stage (P=0.01) and positive uPA expression (P=0.02) were the parameter of overall patients' outcome. However only lymph node status was the significant factor of survival in SCC (P=0.04), this was in line with a study performed on schistosomal associated bladder carcinoma by Ghoneim et al (20). Failure of proving the survival impact of grade, stage and size of tumor in SCC subgroup could be explained by the fact that most of cases studied were of high grade and advanced stage so masking any statistical significance.

From the previous study we can conclude that uPA protein expression was significantly higher in the malignant group than in the control- cystitis- group. Cystitis cases that express uPA should be strictly followed up. Expression of uPA gene is associated with more invasive tumors. Positive uPA expression in the malignant group was significantly correlated with aggressive tumor parameters, high grade, positive lymph node metastasis and advanced tumor stage and the resulting poorer prognosis. Tumor histopathology, grade, tumor stage, lymph node status, vascular invasion and positive uPA protein expression are independent predictors for patients' outcome. Therefore, determination of uPA expression might be useful for predicting malignant bladder carcinoma patients with poor prognosis. Other extended studies should be done on large number of non-malignant cases aiming to know the role of uPA in bladder carcinogenesis.