Abstract P2-03-07: Digital image analysis of Ki67 in hot spots is superior to alternative proliferation associated markers in breast cancer

Abstract

INTRODUCTION Proliferative activity is one of the most important prognostic parameters in cancer. During the pathological examination of breast tumors, it is routinely evaluated by a count of the number of mitoses. Adding immunohistochemical stains of the nuclear protein Ki67 provides extra prognostic and predictive information. However, the currently used methods for both of these evaluations battle imperfections, primarily in reproducibility. In this study, we make an equally broad and detailed evaluation of mitoses, Ki67 and the more recently described Phosphohistone H3 (PHH3) in primary breast cancer using digital image analysis (DIA). Furthermore, we aim to investigate the prognostic and predictive value of proliferation-associated biomarkers in breast cancer stromal cells in relation to patient outcome. MATERIALS AND METHODS Two cohorts of primary breast cancer specimens (total n=297) with clinicopathological data including >10 years survival data, were sectioned and stained for Ki67, PHH3 and pancytokeratin (CKMNF116) and all glass slides were digitally scanned at x20. The DIA software used was the Visiopharm Integrator System (VIS) by Visiopharm A/S, Hoersholm, Denmark. VIS operates by a 'digital fusion' method that automatically excludes non-epithelial tissue restricting the analysis of the biomarkers (Ki67 and PHH3) to CKMNF116 positive cells. Both manual and DIA scores were compared for sensitivity and specificity for the gene expression based Luminal B versus A subtype, for high versus low transcriptomic grade as well as for their prognostic value in terms of Cox regression hazard ratios and breast cancer specific and overall survival. Further, we investigated whether the expression of Ki67 in the tumors' hot spots, invasive edges or as an average across all regions should be assessed for maximum power in relation to these outcomes. In addition, by inverting the DIA algorithm run by the VIS on the same cohorts, the expression of Ki67 and PHH3 was evaluated in the tumor stromal compartment. RESULTS Regardless of tumor region, DIA of Ki67 outperformed the other markers in sensitivity and specificity for gene expression subtypes and transcriptomic grade. In contrast to mitotic counts, tumors with high expression of Ki67 as defined by DIA, had significantly increased hazard ratio for all-cause mortality within 10 years from diagnosis. DIA of Ki67 was superior to manual Ki67 and PHH3 evaluations as well as to mitotic counts in terms of separation of patients with poor versus relatively good survival. Finally, we replaced the manual mitotic counts with DIA of Ki67 in hot spots as the marker for proliferation when determining histological grade. This increased the differences in estimated mean overall survival between the highest and lowest grades and added significantly more prognostic information to the classic Nottingham histological grade. CONCLUSIONS We conclude that digital image analysis of Ki67 in hot spots should be suggested as the marker of choice for proliferative activity in breast cancer. Citation Format: Robertson S, Stålhammar G, Wedlund L, Gholizadeh S, Lippert M, Rantaleinen M, Bergh J, Hartman J. Digital image analysis of Ki67 in hot spots is superior to alternative proliferation associated markers in breast cancer [abstract]. In: Proceedings of the 2017 San Antonio Breast Cancer Symposium; 2017 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2018;78(4 Suppl):Abstract nr P2-03-07.