Computerized image analysis of Ki-67 in ductal breast carcinoma in situ

OBJECTIVE: To develop and determine the staining protocols and computerized image analysis methods that are the most effective combination for perform ing quantitative analysis of Ki-67. STUDY DESIGN: We compared conventional bright-field light microscopy and re fractive optical enhancement methods in combination with various immunodete ction and filter enhancement methods, including immunogold in combination w ith epipolarization refractive optics and enzymatic conversion of chromogen ic substrates in combination with optical filter enhancement. Initial Ki-67 tests were performed on lymph node tissues and cultured human breast cells and then applied to 200 ductal carcinoma in situ (DCIS) samples. DCIS acin i were digitally acquried, and a region of interest was manually outlined i n each one with a digital stylus to include only the cellular component; th en the Ki-67 staining index was quantified by segmentation analysis. RESULTS: Although combining epipolarization analysis with immunohistogold s taining was the most sensitive detection method, nonspecific binding was to o high. The streptavidin-horseradish-peroxide enzymatic conversion of 3,3 ' -diaminobenzidine (DAB) in combination with optical enhancement filters wa s the most effective method tested. Ki-67 stain was associated with dense f ibrillar structures of the nucleoli in tile less intensely staining nuclei and was most intense in paired nuclei. CONCLUSION: The method of measuring Ki-67 expression by DAB staining combin ed with optical enhancement filters and quantification via computer assiste d image analysis techniques produced objective and reproducible results. As such, this method can offer (1) less intraobserver and interobserver varia bility, (2) a digital archival record, and (3) a baseline for digital excha nge of information between studies.