Computational pathology to improve biomarker testing in breast cancer: how close are we?

Abstract

The recent advancements in breast cancer precision medicine have highlighted the urgency for the precise and reproducible characterization of clinically actionable biomarkers. Despite numerous standardization efforts, biomarker testing by conventional methodologies is challenged by several issues such as high inter-observer variabilities, the spatial heterogeneity of biomarkers expression, and technological heterogeneity. In this respect, artificial intelligence-based digital pathology approaches are being increasingly recognized as promising methods for biomarker testing and subsequently improved clinical management. Here, we provide an overview on the most recent advances for artificial intelligence-assisted biomarkers testing in breast cancer, with a particular focus on tumorinfiltrating lymphocytes, programmed death-ligand 1, phosphatidylinositol-3 kinase catalytic alpha, and estrogen receptor 1. Challenges and solutions for this integrative analysis in pathology laboratories are also provided.

Early Breast Cancer Risk Assessment: Integrating Histopathology with Artificial Intelligence

“…[ 56 ]. AI-assisted TIL quantification remains a promising tool when handled by an experienced pathologist; however, forecasting clinical outcomes based on pre-treatment histopathologic images remains a challenging endeavor, hindered by the incomplete comprehension of the tumor immune microenvironment [ 52 , 56 , 57 ].…”

“…However, the performance of IHC is resource-intensive, time-consuming, costly, contingent on specific tissue-handling protocols, and relies on pathologists’ subjective interpretation [ 71 , 72 ]. To address the latter concern, digital image analysis (DIA) has been widely employed in interpreting IHC staining [ 57 , 71 , 72 ]. While image analysis through ML is increasingly utilized across various pathology applications, it has yet to be suggested as a replacement for chemical-based assays in molecular detection [ 71 , 72 ].…”

“…While image analysis through ML is increasingly utilized across various pathology applications, it has yet to be suggested as a replacement for chemical-based assays in molecular detection [ 71 , 72 ]. Ongoing research endeavors seek to integrate both molecular and morphological tumor characteristics, aiming to enhance the prognostic and predictive capabilities of ML methodologies [ 57 ].…”

“…This shift has fundamentally transformed the treatment approach for HER2-low disease, with the proven effectiveness of antibody–drug conjugates (ADCs) serving as a significant addition to the therapeutic arsenal [ 98 , 99 ]. The focus now turns to the importance of precise HER2 evaluation, which can be facilitated by advanced AI technologies [ 27 , 57 , 100 ]. Various methods are employed in HER2 IHC AI algorithms: some of them are based on tumor cells’ segmentation, while others evaluate HER2 membrane staining intensity and patterns [ 101 , 102 , 103 ].…”

Standardized pathology report for HER2 testing in compliance with 2023 ASCO/CAP updates and 2023 ESMO consensus statements on HER2-low breast cancer

“…It is crucial to develop assays that offer more quantitative measurements than IHC to examine these samples and identify potential differential benefits or responses based on quantifiable measures of HER2 protein. In this respect, artificial intelligence (AI) holds significant potential in revolutionizing HER2 testing [ 59 ]. Considering the prevalent discordance in HER2 status between primary and metastatic disease that affects a substantial portion of cases, it's prudent to consider evaluating the HER2-low status using metastatic tissues initially.…”