Aggressive triple negative breast cancers have unique molecular signature on the basis of mitochondrial genetic and functional defects.

Metastatic breast cancer is a leading cause of cancer-related deaths in women worldwide. Patients with triple negative breast cancer (TNBCs), a highly aggressive tumor subtype, have a particularly poor prognosis. Multiple reports demonstrate that altered content of the multicopy mitochondrial genome (mtDNA) in primary breast tumors correlates with poor prognosis. We earlier reported that mtDNA copy number reduction in breast cancer cell lines induces an epithelial-mesenchymal transition associated with metastasis. However, it is unknown whether the breast tumor subtypes (TNBC, Luminal and HER2+) differ in the nature and amount of mitochondrial defects and if mitochondrial defects can be used as a marker to identify tumors at risk for metastasis. By analyzing human primary tumors, cell lines and the TCGA dataset, we demonstrate a high degree of variability in mitochondrial defects among the tumor subtypes and TNBCs, in particular, exhibit higher frequency of mitochondrial defects, including reduced mtDNA content, mtDNA sequence imbalance (mtRNR1:ND4), impaired mitochondrial respiration and metabolic switch to glycolysis which is associated with tumorigenicity. We identified that genes involved in maintenance of mitochondrial structural and functional integrity are differentially expressed in TNBCs compared to non-TNBC tumors. Furthermore, we identified a subset of TNBC tumors that contain lower expression of epithelial splicing regulatory protein (ESRP)-1, typical of metastasizing cells. The overall impact of our findings reported here is that mitochondrial heterogeneity among TNBCs can be used to identify TNBC patients at risk of metastasis and the altered metabolism and metabolic genes can be targeted to improve chemotherapeutic response.

Association of Event-Free and Distant Recurrence-Free Survival With Individual-Level Pathologic Complete Response in Neoadjuvant Treatment of Stages 2 and 3 Breast Cancer: Three-Year Follow-up Analysis for the I-SPY2 Adaptively Randomized Clinical Trial.

Importance: Pathologic complete response (pCR) is a known prognostic biomarker for long-term outcomes. The I-SPY2 trial evaluated if the strength of this clinical association persists in the context of a phase 2 neoadjuvant platform trial. Objective: To evaluate the association of pCR with event-free survival (EFS) and pCR with distant recurrence-free survival (DRFS) in subpopulations of women with high-risk operable breast cancer treated with standard therapy or one of several novel agents. Design, Setting, and Participants: Multicenter platform trial of women with operable clinical stage 2 or 3 breast cancer with no prior surgery or systemic therapy for breast cancer; primary tumors were 2.5 cm or larger. Women with tumors that were ERBB2 negative/hormone receptor (HR) positive with low 70-gene assay score were excluded. Participants were adaptively randomized to one of several different investigational regimens or control therapy within molecular subtypes from March 2010 through 2016. The analysis included participants with follow-up data available as of February 26, 2019. Interventions: Standard-of-care neoadjuvant therapy consisting of taxane treatment with or without (as control) one of several investigational agents or combinations followed by doxorubicin and cyclophosphamide. Main Outcomes and Measures: Pathologic complete response and 3-year EFS and DRFS. Results: Of the 950 participants (median [range] age, 49 [23-77] years), 330 (34.7%) achieved pCR. Three-year EFS and DRFS for patients who achieved pCR were both 95%. Hazard ratios for pCR vs non-pCR were 0.19 for EFS (95% CI, 0.12-0.31) and 0.21 for DRFS (95% CI, 0.13-0.34) and were similar across molecular subtypes, varying from 0.14 to 0.18 for EFS and 0.10 to 0.20 for DRFS. Conclusions and Relevance: The 3-year outcomes from the I-SPY2 trial show that, regardless of subtype and/or treatment regimen, including 9 novel therapeutic combinations, achieving pCR after neoadjuvant therapy implies approximately an 80% reduction in recurrence rate. The goal of the I-SPY2 trial is to rapidly identify investigational therapies that may improve pCR when validated in a phase 3 confirmatory trial. Whether pCR is a validated surrogate in the sense that a therapy that improves pCR rate can be assumed to also improve long-term outcome requires further study. Trial Registration: ClinicalTrials.gov Identifier: NCT01042379.

Mitosis detection in breast cancer pathology images by combining handcrafted and convolutional neural network features.

Breast cancer (BCa) grading plays an important role in predicting disease aggressiveness and patient outcome. A key component of BCa grade is the mitotic count, which involves quantifying the number of cells in the process of dividing (i.e., undergoing mitosis) at a specific point in time. Currently, mitosis counting is done manually by a pathologist looking at multiple high power fields (HPFs) on a glass slide under a microscope, an extremely laborious and time consuming process. The development of computerized systems for automated detection of mitotic nuclei, while highly desirable, is confounded by the highly variable shape and appearance of mitoses. Existing methods use either handcrafted features that capture certain morphological, statistical, or textural attributes of mitoses or features learned with convolutional neural networks (CNN). Although handcrafted features are inspired by the domain and the particular application, the data-driven CNN models tend to be domain agnostic and attempt to learn additional feature bases that cannot be represented through any of the handcrafted features. On the other hand, CNN is computationally more complex and needs a large number of labeled training instances. Since handcrafted features attempt to model domain pertinent attributes and CNN approaches are largely supervised feature generation methods, there is an appeal in attempting to combine these two distinct classes of feature generation strategies to create an integrated set of attributes that can potentially outperform either class of feature extraction strategies individually. We present a cascaded approach for mitosis detection that intelligently combines a CNN model and handcrafted features (morphology, color, and texture features). By employing a light CNN model, the proposed approach is far less demanding computationally, and the cascaded strategy of combining handcrafted features and CNN-derived features enables the possibility of maximizing the performance by leveraging the disconnected feature sets. Evaluation on the public ICPR12 mitosis dataset that has 226 mitoses annotated on 35 HPFs ([Formula: see text] magnification) by several pathologists and 15 testing HPFs yielded an [Formula: see text]-measure of 0.7345. Our approach is accurate, fast, and requires fewer computing resources compared to existent methods, making this feasible for clinical use.