Affiliation:
1. Institute of Informatics of the Faculty of Mathematics and Informatics of Vilnius University
2. Department of Pathology and Forensic Medicine, Institute of Biomedical Sciences of the Faculty of Medicine of Vilnius University
3. Path-Image/BioTiCla, University of Caen, François Baclesse Comprehensive Cancer Center
Abstract
Abstract
Background Immunohistochemistry (IHC) for ER, PR, HER2, and Ki67 is used in breast cancer (BC) pathology to assess tumor properties and predict patient outcomes and therapy responses. Visual scoring of the IHC biomarkers by pathologists, apart from reproducibility issues, does not sufficiently account for the intratumoral heterogeneity (ITH), often a subvisual feature within the tumor tissue. It has been reported that the ITH indicators of IHC biomarker expression can provide independent prognostic value. In this study, we applied digital image analysis (DIA) and computational pathology methods to investigate the prognostic value of IHC ITH indicators in hormone receptor-positive (HR-positive) BC patients.Methods Whole slide images of surgically excised specimens stained for ER, PR, Ki67, and HER2 from 178 patients with a diagnosis of HR-positive invasive ductal carcinoma were used in the study. Digital tumor tissue segmentation and detection of biomarker-positive and negative cells were performed. The DIA-generated data were systematically subsampled by a hexagonal grid to compute Haralick’s texture indicators for ER, PR, Ki67, and HER2. Univariate and multivariable Cox regression analyses were performed to assess the prognostic significance of the IHC and ITH indicators in the context of clinicopathologic variables, including conventional assessment of the IHC results provided by pathologists.Results In multivariable analysis, the ITH of Ki67-positive cells, measured by Haralick’s texture entropy, emerged as an independent prognostic factor associated with worse overall survival (hazard ratio = 11.40, p-value = 0.021). Remarkably, the entropy representing the spatial disarrangement of tumor proliferation outperformed the proliferation rate per se established either by pathology reports or DIA. None of the clinicopathologic variables were selected as independent prognostic features in our dataset.Conclusions These results add to the evidence from previous studies that ITH of IHC biomarkers, in particular, ITH of Ki67 proliferation index, exceeds the informative value of Ki67% per se (both visual and digital) in HR-positive BC. The study further demonstrates the benefits of high-capacity DIA-generated data for quantifying the essentially subvisual ITH properties.
Publisher
Research Square Platform LLC
Reference50 articles.
1. Arzmi MH, P. P. Abdul Majeed A, Muazu Musa R, Mohd Razman MA, Gan H-S, Mohd Khairuddin I, et al. Epidemiology, Detection and Management of Cancer: An Overview. Deep Learning in Cancer Diagnostics: A Feature-based Transfer Learning Evaluation. Singapore: Springer Nature Singapore; 2023. p. 1–7.
2. Understanding breast cancer as a global health concern;Wilkinson L;The British Journal of Radiology,2022
3. Estrogen and Progesterone Receptor Testing in Breast Cancer: American Society of Clinical Oncology/College of American Pathologists Guideline Update;Allison KH;Archives of Pathology & Laboratory Medicine,2020
4. Nielsen TO, Leung SCY, Rimm DL, Dodson A, Acs B, Badve S, et al. Assessment of Ki67 in Breast Cancer: Updated Recommendations From the International Ki67 in Breast Cancer Working Group. JNCI: Journal of the National Cancer Institute. 2020;113(7):808 – 19.
5. Human Epidermal Growth Factor Receptor 2 Testing in Breast Cancer: American Society of Clinical Oncology/College of American Pathologists Clinical Practice Guideline Focused Update;Wolff AC;Arch Pathol Lab Med,2018