Efficient and cost-effective breast cancer risk stratification using whole slide histopathology images

使用全玻片组织病理学图像进行高效且经济的乳腺癌风险分层

基本信息

批准号：
10649978
负责人：
Metin Nafi Gurcan
金额：
$ 19.09万
依托单位：
WAKE FOREST UNIVERSITY HEALTH SCIENCES
依托单位国家：
美国
项目类别：
财政年份：
2023
资助国家：
美国
起止时间：
2023-04-06 至 2025-03-31
项目状态：
未结题

来源：
https://reporter.nih.gov/project-details/10649978
关键词：
Age Attention Biological Assay Biological Markers Biopsy Breast Cancer Patient Breast Cancer Risk Factor Breast Cancer therapy Cancer Prognosis Categories Cell Nucleus Cells Classification Clinical Clinical Research Clinical Trials Consumption Data Developing Countries Diagnosis Diagnostic Disease Duct (organ) structure ERBB2 gene Effectiveness Ensure Epithelium Estrogen Receptors Estrogen receptor positive Estrogens Gene Expression Genes Health Hematoxylin and Eosin Staining Method Histologic Histology Histopathology Human Image Image Analysis Internet Interobserver Variability Learning Malignant Neoplasms Measures Methods Mission Modeling Nature Outcome Patients Performance Play Population Premature Menopause Progesterone Progesterone Receptors Prognosis Public Health Recurrence Recurrence Score Regional Anatomy Research Resource-limited setting Resources Reverse Transcriptase Polymerase Chain Reaction Risk Slide Stains Stratification Time Tissues Training Tumor Tissue Tumor stage United States United States National Institutes of Health Validation Woman breast imaging chemotherapy cohort cost cost effective design detection method diagnostic accuracy erbB-2 Receptor high risk hormone receptor-positive imaging biomarker improved large datasets malignant breast neoplasm mortality novel strategies oncotype optimal treatments overexpression patient stratification personalized care prevent progesterone receptor positive receptor risk stratification rural area side effect tool tumor web-based tool whole slide imaging

项目摘要

Efficient and cost-effective breast cancer risk stratification using whole-slide histopathology images Breast cancer prognosis depends highly on receptor status, as optimal treatment depends on the presence or absence of overexpression of estrogen, progesterone, or HER-2/neu receptors. To prevent over-treating patients with chemotherapy, it is crucial to quantify the risk of recurrence for estrogen receptor (ER) positive (ER+), HER2 negative (HER2-) breast cancer. A common assessment method to meet this need is the Oncotype DX (ODX) Recurrence Score. Unfortunately, ODX and similar gene assays are expensive, time-consuming, and tissue destructive. As an alternative, we propose estimating the ODX recurrence score using routine, ubiquitous, and inexpensive hematoxylin and eosin (H&E) staining of biopsies. There are other efforts to predict ODX recurrence risk from H&E. These automated methods detect histological primitives (e.g., nuclei) often in specific, also automatically detected, anatomical regions (e.g., ducts, tubules, lumen, epithelium, and stroma). Classification is performed into two or three risk categories, often collapsing two categories into one. The performance of these models is promising but still modest. One way to improve the performance of the models is to train on larger datasets; however, annotating larger datasets is challenging. Here, we propose an automated method to predict ODX recurrence risk without annotations. If successful, this method would have a wide range of applications, including but not limited to the availability of an inexpensive, web-based tool to predict ODX in developing countries or rural areas with internet access where standard Oncotype Dx assay would be cost-prohibitive or take too long to obtain. Furthermore, our method would find use in clinical research where valuable tumor tissue could be saved by obtaining correlative research data based on standard H&E-stained slides.

有效且具有成本效益图像乳腺癌预后高度取决于受体状态，因为最佳治疗取决于雌激素，孕酮或HER-2/NEU受体的过表达的存在或不存在。为了防止过度治疗的化学疗法患者，量化的风险至关重要雌激素受体（ER）阳性（ER+），HER2阴性（HER2-）乳腺癌的复发。一个满足这种需求的常见评估方法是Oncotype DX（ODX）复发评分。不幸的是，ODX和类似基因测定很昂贵，耗时和组织破坏性。作为替代方案，我们建议使用例程估算ODX复发评分，活检的无处不在，廉价的苏木精和曙红（H＆E）染色。还有其他预测H＆E的ODX复发风险的努力。这些自动化方法检测组织学原语（例如核）通常以特定的（也自动检测到）的解剖区域（例如，管道，小管，管腔，上皮和基质）。分类分为两个或三个风险类别通常将两个类别崩溃为一类。这些模型的性能是有希望但仍然适中。提高模型性能的一种方法是训练较大的数据集；但是，注释较大的数据集具有挑战性。在这里，我们建议自动化方法可以预测无注释而无需注释。如果成功，这方法将具有广泛的应用，包括但不限于可用性廉价，基于网络的工具，可预测发展中国家或农村地区的ODX 访问标准的ONCOTYPE DX分析将过于良好或花费太长时间才能获得。此外，我们的方法可以在临床研究中找到有价值的肿瘤组织可以通过基于标准H＆E染色幻灯片获得相关研究数据来保存。