Abstract 4853: Comparison of automated and a semi-automated methods for assessing Ki-67 staining.

Abstract

Abstract Human evaluation of immuno-staining is limited by the subjectivity of the human evaluator, is time-consuming and is expensive. Semi-automated means of tissue evaluation have been proposed, but most of these have large subjective components. This study assesses semi-automated and newly-developed fully automated methods of evaluating Ki-67 staining of benign and malignant tissue from radical prostatectomy specimens. A research tissue microarray was constructed from 68 prostatectomy specimens and contained 3 core samples of benign and 3 of malignant tissue from each research subject. Tissue microarray sections were immune-stained for Ki-67. Two technicians (SG & CM) evaluated each TMA core and labeled the nuclei positive for Ki-67 with a green dot, and negative with a red dot. Images were collected and Image-Pro software was used to count the number of green and red dots. In a separate procedure, Image-Pro (IP) software was used to identify nuclei, and grade each nucleus as positive or negative, and the percentage of positive nuclei was calculated. The semi-automated results correlated with each other (0.58, P<0.001). The semi-automated and fully automated results also correlated (0.8, P<0.001, CM vs. IP and 0.59, P<0.001, SG vs. IP). When stratified by tissue type (benign vs. malignant), the semi-automated results and the fully-automated results also correlated (benign tissue: 0.68, P<0.001, CM vs. IP, and 0.80, P<0.001, SG vs. IP; tumor tissue: 0.89, P<0.001, CM vs. IP and 0.47, P<0.001, SG vs. IP). These results suggest that semi-automated and automated analysis of Ki-67 staining of prostate tissue produces comparable results, and automated analysis can be used for prostate cancer research. Citation Format: Rochelle P. Ondracek, James Marshall, Karin Kasza, Christopher Morrison, Simoya Ghajar, Carl Morrison, Diana Mehedint, James L. Mohler. Comparison of automated and a semi-automated methods for assessing Ki-67 staining. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 4853. doi:10.1158/1538-7445.AM2013-4853