The results of colorectal cancer treatment by p53 status - Treatment-specific overview

PURPOSE: Both negative and positive influences of mutant p53 on treatment o utcome hare been reported, and we present here a meta-analysis of published studies where outcome was reported for defined treatment groups. METHODS: We identified articles on the effect of p53 status by treatment modality, e xcluding those not stratified by method of treatment. A common hazard ratio was estimated from studies that reported a multivariate analysis. We also estimated the numbers of patients expressing the endpoint at the mean or me dian follow-up time and calculated a pooled odds ratio. RESULTS: Twenty-eig ht articles were evaluable (23 using immunohistochemistry to detect overexp ression of p53 and 8 using DNA sequencing), for a total of 4,416 patients. For patients treated with surgery only, the immunohistochemistry studies sh owed a significant influence of p53 status on disease-free survival and a m arginally significant influence on overall survival. In the studies using D NA sequencing, by contrast, there was a significant influence of p53 mutati ons on overall survival, but not disease-free survival. For patients treate d with surgery and radiotherapy, the influence of p53 status on disease-fre e survival was either insignificant or marginally significant, depending on test used; there was no influence on overall survival. CONCLUSIONS: Althou gh this pooled analysis of published studies where treatment was accounted for shows that there is a borderline significant hazard associated with p53 overexpression or mutation vs. p53 wild-type, it is unlikely that p53 can be applied in a routine clinical setting along side factors such as T stage , nodal status, and residual tumor, whose prognostic value is much stronger .