Nicolas Elie1, Benoît Plancoulaine, Jehan-Pierre Signolle, Paulette Herlin. 1. Groupe Régional d'Etudes sur le Cancer, Equipe d'Accueil Universitaire 1772, Centre Régional de Lutte Contre le Cancer François Baclesse, route de Lion sur Mer, F-14076 Caen cedex 5, France. n.elie@baclesse.fr
Abstract
BACKGROUND: Quantification of nuclei, immunostained for proliferation or differentiation markers, is widely recommended for prognostication and treatment of numerous solid tumors. However, many problems concerning reproducibility of the results obtained still remain, partially due to the lack of standardization of the method used and to the difficulty of taking into account tumor tissue heterogeneity. Technical solutions exist but they are expensive and time consuming, and their complexity limits their diffusion in routine pathology departments. METHODS: In the present study, we use a simple, fast and inexpensive system of acquisition, a 4,000 dots-per-inch slide scanner, to get a single true color numerical image of the whole histologic section. The analysis is done thanks to fully automatic image processing, allowing computation of the stain proportion on the whole histologic section, as well as in "hot spots." RESULTS: After having precisely fixed settings of the slide scanner, one can easily detect counterstained and immunostained structures, by image analysis. This simple and stable processing can be done under 6 min. CONCLUSIONS: The association of immunohistochemistry, stable acquisition device and computer-assisted image analysis provides an objective, reproducible, and easy to standardize quantification of the nuclear markers, in relation to the total tumor tissue architecture. Copyright 2003 Wiley-Liss, Inc.
BACKGROUND: Quantification of nuclei, immunostained for proliferation or differentiation markers, is widely recommended for prognostication and treatment of numerous solid tumors. However, many problems concerning reproducibility of the results obtained still remain, partially due to the lack of standardization of the method used and to the difficulty of taking into account tumor tissue heterogeneity. Technical solutions exist but they are expensive and time consuming, and their complexity limits their diffusion in routine pathology departments. METHODS: In the present study, we use a simple, fast and inexpensive system of acquisition, a 4,000 dots-per-inch slide scanner, to get a single true color numerical image of the whole histologic section. The analysis is done thanks to fully automatic image processing, allowing computation of the stain proportion on the whole histologic section, as well as in "hot spots." RESULTS: After having precisely fixed settings of the slide scanner, one can easily detect counterstained and immunostained structures, by image analysis. This simple and stable processing can be done under 6 min. CONCLUSIONS: The association of immunohistochemistry, stable acquisition device and computer-assisted image analysis provides an objective, reproducible, and easy to standardize quantification of the nuclear markers, in relation to the total tumor tissue architecture. Copyright 2003 Wiley-Liss, Inc.
Authors: W Kocha; J Maroun; H Kennecke; C Law; P Metrakos; J F Ouellet; R Reid; C Rowsell; A Shah; S Singh; S Van Uum; R Wong Journal: Curr Oncol Date: 2010-06 Impact factor: 3.677