Correlation between laminin and type IV collagen distribution in breast carcinomas, and estrogen receptors expression, lymph node and vascular involvement

Abstract

The laminin (Lam) and type IV collagen (Coll IV) and estrogen receptor (ER) immunodetections were assessed in a large series of 400 human breast carcinomas. In all the cases the patient's age, the tumor size, the histological type and grade, the presence or the absence of axillary lymph node metastasis and of vascular invasion in tumor borders, and ER tumor content were recorded. Monoclonal anti-Lam, anti-Coll IV were applied with the avidin-biotin-peroxidase complex and monoclonal anti ER with peroxidase anti-peroxidase complex, on frozen sections. A computerized system of image analysis referred to as SAMBA (TITN) with specific software for tissue sections analysis permitted a multiparametric quantitative analysis of immunostained surfaces. With this system, in each tumor, the cellularity, the percentage of Lam, Coll IV and receptor positive surfaces versus the total cell surface and versus the epithelial (keratin positive) surface, the integrated optical density IOD histograms were obtained and correlated to morphometrical and standard histological data. From this study, it was shown that: (1) Lam and Coll IV immunostained epithelial basement membranes in carcinomas were correlated to the presence of estrogen and progesterone receptor antigenic sites within the tumors, with a significant decrease of the positive staining in ER-ICA negative tumors in comparison to ER-ICA/PR-ICA positive tumors. (2) The combined densitometric and morphometric evaluation demonstrated a decrease of Lam and Coll IV immunostaining in malignant tumors, correlated to (i) the presence of peritumorous vascular invasion and (ii) keratin positive cells in bone marrow (iii) axillary lymph node involvement. It is concluded that the variations in Lam and Coll IV antigens distribution may be relevant indicators of tumor metastatic potential in breast carcinomas and that computerized image analysis enables the standardization of the evaluation antigens distribution.