Bcl-2 automated quantitative immunocytochemical assays in breast carcinomas: correlation with 10-year follow-up

ArticleinJournal of Clinical Oncology 16(6):2025-31 · July 1998with7 Reads
Impact Factor: 18.43 · Source: PubMed


    bcl-2 protein is detectable in human cancers and may be involved in the response to antineoplastic drugs or endocrine therapy in breast carcinomas. In a previous study, we had developed optimal technical conditions for bcl-2 immunodetection. The aim of the present report was to determine the prognostic significance of bcl-2 expression in breast carinomas by the use of a similar immunocytochemical procedure.
    bcl-2 immunocytochemical assays were performed on frozen sections by automated immunoperoxidase technique (Ventana) and computer-assisted analysis of digitized colored microscopic images (SAMBA) in a series of 170 breast carcinomas. The results of automated quantitative immunocytochemical assays were correlated with patient follow-up (120 months).
    Intense bcl-2 immunocytochemical expression in tumors (cutpoint, 15%) significantly correlated with longer disease-free survival and longer recurrence-free survival in the entire cohort of patients (P = .028 and P = .035, respectively) and also in node-negative subgroups of patients (P = .028 and P = .01; Kaplan-Meier long-rank test; NCSS 6.0.1 software). But bcl-2 immunostained surfaces (cutpoint, 15%) did not correlate with overall survival. In multivariate analysis (proportional hazards regression, Cox model), bcl-2 prognostic significance in terms of disease-free survival was only independent of the tumor size and grade and histoprognostic index (Nottingham prognostic index [NPI]).
    bcl-2 immunohistochemical expression is a significant indicator of favorable outcome only in terms of disease-free and local recurrence-free survival. However, bcl-2 expression in tumors is an independent weakly prognostic indicator in breast carcinomas. bcl-2 immunodetection assessed in optimal technical conditions (frozen samples, automation, quantitative analysis, scatter diagram cutoffs) may have some limited practical clinical relevance for the management of patients with breast carcinomas.