Tumor Neoangiogenesis by CD31 and CD105 Expression Evaluation in Breast Carcinoma Tissue Microarrays

• Published in: Clinical cancer research 2004-09, Vol.10 (17), p.5815-5819
• Main Authors: CHARPIN, Colette, DALES, Jean-Philippe, GARCIA, Stéphane, CARPENTIER, Séverine, DJEMLI, Amina, ANDRAC, Lucile, LAVAUT, Marie-Noëlle, ALLASIA, Claude, BONNIER, Pascal
• Format: Article
• Language: English
• Subjects: Adenocarcinoma - blood supply; Adenocarcinoma - diagnosis; Adult; Aged; Aged, 80 and over; Analysis of Variance; Antigens, CD; Antineoplastic agents; Biological and medical sciences; Breast Neoplasms - blood supply; Breast Neoplasms - diagnosis; Breast Neoplasms - metabolism; Endoglin; Female; Follow-Up Studies; Humans; Image Processing, Computer-Assisted; Immunoenzyme Techniques; Medical sciences; Middle Aged; Neovascularization, Pathologic - diagnosis; Neovascularization, Pathologic - metabolism; Oligonucleotide Array Sequence Analysis; Paraffin Embedding; Pharmacology. Drug treatments; Platelet Endothelial Cell Adhesion Molecule-1 - metabolism; Prognosis; Receptors, Cell Surface; Survival Rate; Tumors; Vascular Cell Adhesion Molecule-1 - metabolism
• ISSN: 1078-0432; 1557-3265
• DOI: 10.1158/1078-0432.CCR-04-0021

Purpose: The aim of the study was to evaluate CD31 and CD105 immunohistochemical expressions in tissue microarrays from 360 breast carcinomas. Study design: Computerized (ACIS/Chromavision) assisted image analysis was performed to compare immunoreactions in tissue microarrays with those in current paraffin and frozen sections. We also aimed to determine the CD105 and CD31 prognostic significance and relevance in routine practice by correlating results of immunodetections with patients’ ( n = 360) outcome (14.3-year follow-up). Results: The results show ( a ) that in tissue microarrays, the CD31 and CD105 expression quantified by image analysis device did not correlate with the measurements assessed on routine paraffin sections; ( b ) that CD105 expression is endowed of a prognostic significance in paraffin sections in terms of overall survival ( P < 0.01), whereas in contrast, CD31 on paraffin sections did not correlate with patients overall survival; ( c ) that semiquantitative analysis of CD105 expression correlated with the image analysis measurements in frozen sections (ρ = 0.671, P < 0.01) and paraffin (ρ = 0.824, P < 0.01) sections. However, paraffin sections were less immunostained than frozen ones. Conclusions: It is concluded ( a ) that CD105 may be suitable in paraffin sections to evaluated neoangiogenesis; and ( b ) that tissue microarrays are not suitable substrates for neoangiogenesis evaluation as a prognostic indicator in breast carcinomas, in contrast to current tissue sections.