Caveolin-1 as prognostic factor of disease recurrence and survival in patients treated with radical cystectomy for bladder cancer.

PURPOSE: Overexpression of Caveolin-1 has been associated with cancer growth, migration, and metastases in several malignancies, but only few data are available on its role in bladder cancer (BCa). The aim of this study is to validate Caveolin-1 as a prognosticator of recurrence-free survival (RFS), overall survival (OS) and cancer-specific survival (CSS) in a large cohort of patients treated with radical cystectomy (RC) for BCa.

METHODS: Caveolin-1 expression was evaluated by immunochemistry on a tissue microarray from 424 patients treated with RC for UCB at a single institution. Caveolin-1 was considered overexpressed when at least 50% of the tumor cells stained positively. Univariable and multivariable Cox proportional hazards regression models were used to assess the association of Caveolin-1 expression with RFS, OS, and CSS.

RESULTS: Overexpression of Caveolin-1 was observed in 116 (27.4%) patients and was associated with lymph node metastasis (P = 0.003). Median follow-up for patients alive at last follow-up was 129 months (interquartile range [IQR]: 82-178). Patients with overexpression of Caveolin-1 had significant worse RFS, OS, and CSS compared to those with normal expression (log-rank test, P = 0.008, P = 0.001, and P = 0.005, respectively). At multivariable analyses that adjusted for the effects of standard clinicopathologic features, Caveolin-1 remained associated with OS (hazard ratio = 1.47, P = 0.002) and CSS (hazard ratio = 1.42, P = 0.03). Conversely, no association with RFS was found (P = 0.1). Addition of Caveolin-1 in a model for prediction of survival did not improve the accuracy of the prognostic model. Actually, C-index did not differ among models with or without Caveolin-1 (0.72 for a model predicting RFS, 0.65 for OS, and 0.71 for CSS).

CONCLUSIONS: Caveolin-1 is overexpressed in one-third of patients with BCa treated with RC. Overexpression of Caveolin-1 is significantly associated with OS and CSS, but not with RFS, in patients with BCa treated with RC. However, it is not clinically useful as it does not improve upon the predictive accuracy of survival achieved by pathologic variables alone.