The tumour suppressor RhoB is an independent prognostic factor for metastasis in urinary bladder cancer.

Apostolos Zaravinos1, Dimitrios Volanis1,2, Ioannis Karyotis2, Ioannis Boulalas1,2, George I. Lambrou3, Vassilis Zoumpourlis4, Dimitris Delakas2, Demetrios A. Spandidos1.

1, Laboratory of Virology, Medical School, University of Crete, 71110 Heraklion, Crete, Greece.2, Department of Urology, Asklipieio General Hospital, 16673 Voula, Athens, Greece.

3, 1st Department of Pediatrics, Choremeio Research Laboratory, University of Athens, 11527 Athens, Greece.4, Unit of Biomedical Applications, Institute of Biological Research and Biotechnology, National Hellenic

Research Foundation, 48 Vas. Constantinou Ave, 116 35 Athens, Greece.

A total of 77 paired samples, consisting of tumour (BC) and normal urothelium, were obtained from 77 Greek patients with TCC of the urinary bladder and studied to determine the expression of RhoA, RhoB, RhoC, Cdc42 and Rac1. The Kaplan-Meier method was used to estimate survival as a function of time, and survival differences were assessed by the Log-rank test. Logistic regression (univariate and multivariate) analysis was performed to determine the potential predictors of survival, recurrence and metastasis. Statistical significance was set at the 95% level (p<0.05). All statistical analyses were performed using SPSS 16.0 (SPSS, Chicago, IL) software.

Urinary bladder cancer is the most common malignancy of the urinary tract, responsible for significant mortality and morbidity worldwide. Almost all bladder cancers are carcinomas, arising from the transitional epithelium. At presentation, 75-85% of tumours are restricted to the mucosa, or invade the lamina propria mucosae. Over 60% of the superficial tumours recur at least once and progress to less differentiated or invasive neoplasms. The most useful prognostic parameters are tumour grade, stage, size, prior recurrence rate and the synchronous presence of carcinoma in situ (CIS). Rho members may affect the process of tumourigenesis either by over-expression of some members of the family with oncogenic activity or by down-modulation of other members with suggested tumour suppressor activity.

The cases whose tumours exhibited increased levels of RhoB mRNA expression exhibited worse overall and cancer-specific survival rates, than those expressing decreased RhoB mRNA levels. Moreover, those cases whose tumours exhibited high Cdc42 mRNA expression showed a worse overall survival rate than those expressing low Cdc42 levels. High RhoC levels tended to correlate with better survival. Although univariate analysis, using the Cox proportional hazards model, showed that RhoB has a tendency for being an independent prognostic factor for overall survival (p=0.086), both univariate and multivariate analysis for RhoA, RhoC, Rac1 and Cdc42 did not exhibit the same tendency. Moreover, no gene was identified as independent prognostic factor for tumour recurrence. Finally, both univariate and multivariate analysis identified RhoB independent prognostic factors for metastasis (p=0.012 and p=0.050, respectively).

Our results confirm a tumour suppressor role for RhoB in bladder cancer, opposing the positive functions of RhoA and RhoC. Moreover, our analysis identified that RhoB could be used as an independent prognostic factor for metastasis.

Variable Relative risk 95% Confidence Interval P value

Overall Survival RhoA 2.381 0.750-7.562 0.141RhoB 0.381 0.126-1.148 0.086RhoC 0.711 0.241-2.094 0.536Rac1 1.964 0.675-5.715 0.215Cdc42 0.442 0.151-1.292 0.136

Recurrence RhoA 1.351 0.500-3.649 0.553RhoB 0.518 0.198-1.351 0.179RhoC 1.426 0.539-3.772 0.475Rac1 1.562 0.598-4.082 0.362Cdc42 0.693 0.269-1.788 0.449

Metastasis RhoA 0.559 0.179-1.746 0.317RhoB 0.232 0.074-0.729 0.012RhoC 0.477 0.161-1.413 0.182Rac1 0.900 0.318-2.551 0.843Cdc42 0.434 0.151-1.252 0.123

Variable Relative risk 95% Confidence Interval P value

Overall Survival RhoB 0.308 0.078-1.211 0.092Rac1 3.194 0.827-12.343 0.092

Recurrence RhoC 1.742 0.633-4.795 0.282

Metastasis RhoB 0.317 0.097-1.035 0.050

Logistic regression analysis [(A) univariate and (B) multivariate] was utilized to assess the potential prognostic factors in overall survival, tumour recurrence and metastasis.

(A)

(B)