Early Breast Cancer Risk Assessment: Integrating Histopathology with Artificial Intelligence

Abstract

Effective risk assessment in early breast cancer is essential for informed clinical decision-making, yet consensus on defining risk categories remains challenging. This paper explores evolving approaches in risk stratification, encompassing histopathological, immunohistochemical, and molecular biomarkers alongside cutting-edge artificial intelligence (AI) techniques. Leveraging machine learning, deep learning, and convolutional neural networks, AI is reshaping predictive algorithms for recurrence risk, thereby revolutionizing diagnostic accuracy and treatment planning. Beyond detection, AI applications extend to histological subtyping, grading, lymph node assessment, and molecular feature identification, fostering personalized therapy decisions. With rising cancer rates, it is crucial to implement AI to accelerate breakthroughs in clinical practice, benefiting both patients and healthcare providers. However, it is important to recognize that while AI offers powerful automation and analysis tools, it lacks the nuanced understanding, clinical context, and ethical considerations inherent to human pathologists in patient care. Hence, the successful integration of AI into clinical practice demands collaborative efforts between medical experts and computational pathologists to optimize patient outcomes.

Keywords: artificial intelligence; biomarkers; breast cancer; deep learning; early breast cancer; pathology; predictive algorithms; risk stratification.

Figure 1

Established and evolving AI-developed approaches of EBC risk definition, involving various biomarkers’ analysis and clinicopathologic features. Traditional histopathological characteristics, detected on H&E staining, form the core of the pathology report, providing essential prognostic information: tumor identification and quantification, tumor size, lymph node involvement, histological grade (according to the Nottingham system), lymphovascular invasion, and sentinel lymph node status. Some pathologists may report TILs, although current recommendations do not suggest basing therapeutic strategies on this biomarker IHC assessment of the hormone receptor and HER2 status (according to ASCO/CAP guidelines), and the Ki67 proliferation index is essential to assign BCs to the luminal/non-luminal molecular classification and to guide treatment choices with both prognostic and predictive implications. In cases suggestive of hereditary BC syndrome, HRD and BRCA1/2 testing is recommended. The developing landscape of AI-based DL algorithms involves the creation of neural networks, capable of predicting IHC status on H&E slides without an actual IHC staining, followed by genomic status and therapy response prediction, risk assessment, and improved patient prognostic stratification. Abbreviations: AI, artificial intelligence; BC, breast cancer; EBC, early breast cancer; DL, deep learning; H&E, hematoxylin and eosin; IHC, immunohistochemistry; pT, primary tumor size; pN, regional lymph node involvement; G, histological grade; LVI, lymphovascular invasion; snLN, sentinel lymph node; TILs, tumor-infiltrating lymphocytes; ER, estrogen receptor; PR, progesterone receptor; HRD, homologous recombination deficiency.