p53 immunohistochemistry in bladder cancer-a new approach to an old question

Objective: For nearly 20 years, the putative prognostic value of P53 immunohistochemistry in bladder cancer has been controversially discussed, and key questions are still unanswered. The aim of this article was to elucidate the different findings using the new concept of a combined analysis of raw data from previously published material. Materials and methods: Twenty-six of 38 study centers approached contributed patient data sets according to the protocol requirements; 3,421 patients with bladder cancer from 25 centers are included in the further analysis. The entire study group (mean age: 66.5 years) comprised 2,697 males (78.8 %) and 719 females. For 2,298 patients (68%) with non-muscle-invasive tumors Ta (1314), TIS (37), and T1 (947) (38.9, 1.1, and 28%, respectively) a median survival time of 130 months was calculated. The remaining 1,082 patients (32%) had advanced tumors (>T2) and a median survival time of 26 months. Of the 1,241 patients who have died, 744 patients died from bladder cancer and another 497 patients from intercurrent disease. Results: With regard to gender, age, and tumor stage, the patients included reflected a normal bladder cancer population. Statistical analysis revealed a highly significant correlation between P53 positivity vs. tumor grade and tumor stage. Uni- and multivariate analyses showed that P53 positivity was significantly correlated with tumor progression (as defined by the different centers) in T1 disease (P < 0.001) as well as in advanced bladder cancer (P < 0.001), but not in Ta tumors. Conclusions: P53 immunohistochemistry appears to be predictive in high grade bladder cancer, however, the magnitude of this association varies among tumor stages. The results of this trial demonstrate the relevance of sufficient study size, provide a basis to define suitable patient populations, and allow an estimation of an adequate size of study cohorts for prospective trials. It also demonstrates the importance of common recommendations for evaluation and reporting of marker studies. Furthermore, this initiative demonstrates the strong will of a lame number of investigators to contribute to combined efforts in order to establish pathways for standardization of marker development and clinical use. (c) 2010 Elsevier Inc. All rights reserved.