BACKGROUND: The endogenous lectin galectin-3 can regulate cell adhesion and proliferation in vitro, thus prompting the examination of its clinical relevance in breast cancer. MATERIALS AND METHODS: Immunohistochemical processing of tissue sections (n = 273; drop-out rate 20.4%) was used for the assessment of galectin-3 expression. Cytoplasmic/nuclear staining and presence in the tumor stroma were analyzed in human breast cancer patients. RESULTS: A weak correlation with positive steroid receptor status was revealed for cytoplasmic positivity. Nuclear staining was correlated to the lobular type of invasive carcinoma, and tumor stroma expression to high-grade malignancy. Multiple testing of cut-off points to divide the cases into groups based on different levels of immunopositivity combined with univariate Kaplan-Meier survival analysis and computations following the multivariate Cox regression model disclosed no prognostic correlation to either cytoplasmic or nuclear expression of galectin-3. The presence of galectin-3 in the stroma, however, indicated an unfavorable prognosis. Prediction of overall survival was feasible using a model consisting of stage and c-erbB2 status. CONCLUSION: These data signify that caution should be exercised in extrapolating from the anti-apoptotic/prometastatic activity of galectin-3 in model systems to the clinical situation.
BACKGROUND: The endogenous lectin galectin-3 can regulate cell adhesion and proliferation in vitro, thus prompting the examination of its clinical relevance in breast cancer. MATERIALS AND METHODS: Immunohistochemical processing of tissue sections (n = 273; drop-out rate 20.4%) was used for the assessment of galectin-3 expression. Cytoplasmic/nuclear staining and presence in the tumor stroma were analyzed in humanbreast cancerpatients. RESULTS: A weak correlation with positive steroid receptor status was revealed for cytoplasmic positivity. Nuclear staining was correlated to the lobular type of invasive carcinoma, and tumor stroma expression to high-grade malignancy. Multiple testing of cut-off points to divide the cases into groups based on different levels of immunopositivity combined with univariate Kaplan-Meier survival analysis and computations following the multivariate Cox regression model disclosed no prognostic correlation to either cytoplasmic or nuclear expression of galectin-3. The presence of galectin-3 in the stroma, however, indicated an unfavorable prognosis. Prediction of overall survival was feasible using a model consisting of stage and c-erbB2 status. CONCLUSION: These data signify that caution should be exercised in extrapolating from the anti-apoptotic/prometastatic activity of galectin-3 in model systems to the clinical situation.
Authors: Jessica R Newton-Northup; Marie T Dickerson; Lixin Ma; Cynthia L Besch-Williford; Susan L Deutscher Journal: Clin Exp Metastasis Date: 2012-08-01 Impact factor: 5.150