Intense cytoplasmic ezrin immunoreactivity predicts poor survival in colorectal cancer

Autor: Kari Syrjänen, Riyad Bendardaf, Seppo Pyrhönen, Suvi Heikkilä, Adam Elzagheid, Olli Carpén, Abdelbaset Buhmeida, Antti Vaheri, Eija Korkeila
Rok vydání: 2008
Předmět:
Zdroj: Human pathology. 39(12)
ISSN: 1532-8392
Popis: Ezrin is a membrane-cytoskeleton anchor, which, in experimental models, regulates tumor cell invasion and metastatic ability. We carried out immunohistochemical analysis of ezrin in 74 advanced colorectal cancer patients and correlated it to clinicopathologic variables and disease outcome. In contrast to the predominantly membraneous immunoreactivity of normal colorectal epithelium, ezrin expression in the colorectal cells was typically cytoplasmic. Altogether, 16.2% (12/74) of the tumors showed negative/weak ezrin staining, 35.1% (26/74) had moderate staining, and 48.6% (36/74) had intense staining. The expression was more intense in colon than in rectal carcinomas (P = .003). Increased ezrin expression was associated with adverse outcome, that is, shorter disease-specific survival; 48.3 months and 36.6 months for negative-weak versus intense expression (P = .041) as well as shorter survival with metastases at 36 months (P = .030); the metastases(36) rates in ezrin(neg/weak), ezrin(moderate), ezrin(intense) are 58.3%, 25.0%, and 18.4%, respectively. In univariate survival analysis, dichotomized (negative/weak versus moderate/strong) ezrin expression significantly predicted both the 5-year disease specific survival (P = .035) and 5-year metastases (P = .018) but lost this predictive power in multivariate (Cox) analysis. High ezrin expression was also related to high E-cadherin (cytoplasmic) expression, DNA aneuploidy, and high thymidylate synthase expression (P = .046, P = .042, P = .046, respectively). These results suggest that ezrin may play a role in colorectal cancer progression and that ezrin expression might provide clinically valuable information in predicting the biological behavior of colorectal cancer.
Databáze: OpenAIRE