Fiche publication


Date publication

août 2005

Auteurs

Membres identifiés du Cancéropôle Est :
Dr CHAPUSOT Caroline , Pr MARTIN Laurent


Tous les auteurs :
Ponnelle T, Chapusot C, Martin L, Bouvier AM, Plenchette S, Faivre J, Solary E, Piard F

Résumé

PURPOSE: The present study was designed to determine whether the nuclear or cytoplasmic expression of survivin, was related to clinicopathological parameters and survival in sporadic colon carcinomas. METHODS: Western blotting of cell fractions and immunocytochemical methodology were used in five human colon cancer cell lines. Immunohistochemical study was performed in formalin-fixed paraffin-embedded section from 46 patients with sporadic colorectal adenocarcinomas with a polyclonal antibody directed against survivin. Apoptotic index was evaluated by using the M30 antibody. Survival rates were estimated by the Kaplan-Meier method and compared using the log-rank test. Multivariate survival analysis was performed by the Cox proportional hazards model. RESULTS: Western blotting and immunocytochemistry analyses confirmed that survivin could be detected both in the nucleus and the cytoplasm. Immunohistochemical analysis demonstrated that 39% of tumours expressed survivin in the nucleus and 41% in the cytoplasm. No relationship was observed between survivin expression and clinicopathological features. Unexpectedly, the apoptotic index appeared to be linked with high survivin nuclear expression. Overall, 3-year observed survival rate was 73% in patients with cytoplasmic survivin expression versus 48% for negative expression (P = 0.14). Survival was 72% versus 50% for positive nuclear survivin expression versus negative (P = 0.16). After adjustment for age and stage, cytoplasmic survivin expression was a significant prognostic factor. A high level of expression was associated to a better survival: RR = 0.35 [0.13-0.98], P = 0.045. CONCLUSION: These results indicate that the analysis of the subcellular expression of survivin is a determining factor to define the prognostic value. Its evaluation, using a polyclonal antibody, might help clinicians in the stratification of patients with colorectal cancer.

Référence

J Cancer Res Clin Oncol. 2005 Aug;131(8):504-10