p53 immunohistochemistry in bladder cancer. Combined analysis: a way to go?

Seven years after the initial reports of the potential prognostic value of p53 immunohistochemistry in bladder cancer, key questions remain unanswered. This is mainly due to conflicting results obtained through multiple retrospective trials mostly performed with small numbers of poorly defined patients. The aim of this article was to investigate the feasibility of a combined analysis of previously published data. Based on a MEDLINE search in 1997, papers reporting on the prognostic role of p53 alterations were identified and the respective groups were invited to participate. Twenty-six of 38 study centers approached contributed patient data sets according to the protocol requirements. Acceptance of the initiative was similar in Europe, North America, and Australasia. A total of 3,421 patients with bladder cancer from 25 centers are included in the further analysis. With regard to gender, age distribution, and tumor stage the patients contributed are comparable with a normal bladder cancer population. Considering tumor grade, G2 tumors appear to be over- represented while the amount of G1 tumors is smaller than expected. The reason for this is probably related to the low rate of p53 accumulation in G1 tumors. No significant differences between those groups initially approached and those that finally cooperated were noted. Therefore, it is concluded that a combined study is feasible, and the patient data collected are representative for the population presented in the published literature. The final analysis of the data is the subject of current investigation.