Deep learning models for histologic grading of breast cancer and association with disease prognosis.

Histologic grading of breast cancer involves review and scoring of three well-established morphologic features: mitotic count, nuclear pleomorphism, and tubule formation. Taken together, these features form the basis of the Nottingham Grading System which is used to inform breast cancer characterization and prognosis. In this study, we develop deep learning models to perform histologic scoring of all three components using digitized hematoxylin and eosin-stained slides containing invasive breast carcinoma.

Determining breast cancer biomarker status and associated morphological features using deep learning.

Background: Breast cancer management depends on biomarkers including estrogen receptor, progesterone receptor, and human epidermal growth factor receptor 2 (ER/PR/HER2). Though existing scoring systems are widely used and well-validated, they can involve costly preparation and variable interpretation. Additionally, discordances between histology and expected biomarker findings can prompt repeat testing to address biological, interpretative, or technical reasons for unexpected results.

Metaplastic carcinomas of the breast without evidence of epithelial differentiation: a diagnostic approach for management.

Aims: Although rare, malignant sarcomatoid breast tumours without evidence of epithelial differentiation comprise a diagnostic challenge with management implications. Earlier studies have generally considered these to be primary breast sarcomas; however, supporting evidence is lacking and management remains variable. This study aimed to provide an evidence-based approach to improve the consistency of diagnosis and management for such cases.

Prosigna® breast cancer assay: histopathologic correlation, development, and assessment of size, nodal status, Ki-67 (SiNK™) index for breast cancer prognosis.

The Prosigna® assay is a United States Food and Drug Administration (US-FDA) cleared molecular test for prognostic use in hormone receptor-positive stage I/II breast cancer in postmenopausal women. We analyzed histopathologic features of 79 cases with Prosigna® assay results and found a significant correlation between tumor size, grade, and Ki-67 labeling index with Prosigna® score (0-40, 41-60, and 61-100) and Prosigna® risk categories. Since the Prosigna® risk stratification is influenced by lymph node status, we designed an index that included lymph node status and the two most correlated variables (size and Ki-67 labeling index).

The healthcare value of the Magee Decision Algorithm™: use of Magee Equations™ and mitosis score to safely forgo molecular testing in breast cancer.

Magee Equations™ are multivariable models that can estimate oncotype DX® Recurrence Score, and Magee Equation 3 has been shown to have chemopredictive value in the neoadjuvant setting as a standalone test. The current study tests the accuracy of Magee Decision Algorithm™ using a large in-house database. According to the algorithm, if all Magee Equation scores are <18, or 18-25 with a mitosis score of 1, then oncotype testing is not required as the actual oncotype recurrence score is expected to be ≤25 (labeled "do not send").