Digital image analysis for Ki67 in breast cancer
Ki67 is a proliferation marker important for classification of malignant tumours such as breast carcinoma. However, manual counting and calculation of Ki67 proliferation index is laborious and prone to inter-observer variance. Additionally, staining platform and choice of monoclonal antibody clone may impact intensity of staining and, in turn, the Ki67 Proliferation Index.
Recently, a computerized algorithm that enables virtual alignment of two consecutive slides stained for pan-cytokeratin and Ki67 has been developed. Software analysis (Virtual Double Staining – VDS) of this image enables exclusion of stromal cells and calculation of Ki67 proliferation indices in tumour cells only. This presentation will discuss the results of experiments validating the usability of this method. Additionally, results obtained by the applied algorithm in experiments examining the impact of antibody clone, format and staining platform on staining quality and proliferation index will be presented. Finally, a new study examining cell cultures using digital image analysis as potential IHC sensitivity indicators will be presented.
Rasmus Røge is Resident in Pathology, Aalborg University Hospital. He is a PhD-student at Aalborg University with his PhD-thesis focusing on validation of Digital Image Analysis algorithm for routine pathology practise. In addition working as scheme organiser of the Nordic Immunohistochemical Quality Control (NordiQC) scheme and assessor in the General, Breast Cancer and Companion diagnostic modules. Fields of Research: Quality Assurance in Immunohistochemistry, Proliferation Markers and Digital Image Analysis.