Correlation between breast cancer subtypes determined by immunohistochemistry and n-COUNTER PAM50 assay: a real-world study.

PURPOSE: Molecular subtyping based on gene expression profiling (i.e., PAM50 assay) aids in determining the prognosis and treatment of breast cancer (BC), particularly in hormone receptor (HR)-positive/human epidermal growth factor receptor 2 (HER2)-negative tumors, where luminal A and B subtypes have different prognoses and treatments. Several surrogate classifications have been proposed for distinguishing between the luminal A and B subtypes. This study determines the accuracy of local immunohistochemistry (IHC) techniques for classifying HR-positive/HER2-negative (HR+/HER2-) tumors according to intrinsic subtypes using the nCOUNTER PAM50 assay as reference and the HR status definition according the ASCO/CAP recommendations. METHODS: Molecular subtypes resulting from nCOUNTER PAM50 performed in our laboratory between 2014 and 2020 were correlated with three different proxy surrogates proposed in the literature based on ER, PR, HER2, and Ki67 expression with different cut-off values. Concordance was measured using the level of agreement and kappa statistics. RESULTS: From 1049 samples with the nCOUNTER test, 679 and 350 were luminal A and B subtypes, respectively. Only a poor-to-fair correlation was observed between the three proxy surrogates and real genomic subtypes as determined by nCOUNTER PAM50. Moreover, 5-11% and 18-36% of the nCOUNTER PAM50 luminal B and A tumors were classified as luminal A and B, respectively, by these surrogates. CONCLUSION: The concordance between luminal subtypes determined by three different IHC-based classifiers and the nCOUNTER PAM50 assay was suboptimal. Thus, a significant proportion of luminal A and B tumors as determined by the surrogate classifiers could be undertreated or over-treated.

Breast pectoral muscle segmentation in mammograms using a modified holistically-nested edge detection network.

This paper presents a method for automatic breast pectoral muscle segmentation in mediolateral oblique mammograms using a Convolutional Neural Network (CNN) inspired by the Holistically-nested Edge Detection (HED) network. Most of the existing methods in the literature are based on hand-crafted models such as straight-line, curve-based techniques or a combination of both. Unfortunately, such models are insufficient when dealing with complex shape variations of the pectoral muscle boundary and when the boundary is unclear due to overlapping breast tissue. To compensate for these issues, we propose a neural network framework that incorporates multi-scale and multi-level learning, capable of learning complex hierarchical features to resolve spatial ambiguity in estimating the pectoral muscle boundary. For this purpose, we modified the HED network architecture to specifically find 'contour-like' objects in mammograms. The proposed framework produced a probability map that can be used to estimate the initial pectoral muscle boundary. Subsequently, we process these maps by extracting morphological properties to find the actual pectoral muscle boundary. Finally, we developed two different post-processing steps to find the actual pectoral muscle boundary. Quantitative evaluation results show that the proposed method is comparable with alternative state-of-the-art methods producing on average values of 94.8 ±â€¯8.5% and 97.5 ±â€¯6.3% for the Jaccard and Dice similarity metrics, respectively, across four different databases.

Concordance of Genomic Variants in Matched Primary Breast Cancer, Metastatic Tumor, and Circulating Tumor DNA: The MIRROR Study.

PURPOSE: Genetic heterogeneity between primary tumors and their metastatic lesions has been documented in several breast cancer studies. However, the selection of therapy for patients with metastatic breast cancer and the search for biomarkers for targeted therapy are often based on findings from the primary tumor, mainly because of the difficulty of distant metastasis core biopsies. New methods for monitoring genomic changes in metastatic breast cancer are needed (ie, circulating tumor DNA [ctDNA] genomic analysis). The objectives of this study were to assess the concordance of genomic variants between primary and metastatic tumor tissues and the sensitivity of plasma ctDNA analysis to identify variants detected in tumor biopsies. PATIENTS AND METHODS: Next-generation sequencing technology was used to assess the genomic mutation profile of a panel of 54 cancer genes in matched samples of primary tumor, metastatic tumor, and plasma from 40 patients with metastatic breast cancer. RESULTS: Using Ion Torrent technology (ThermoFisher Scientific, Waltham, MA), we identified 110 variants that were common to the primary and metastatic tumors. ctDNA analysis had a sensitivity of 0.972 in detecting variants present in both primary and metastatic tissues. In addition, we identified 13 variants in metastatic tissue and ctDNA not present in primary tumor. CONCLUSION: We identified genomic variants present in metastatic biopsies and plasma ctDNA that were not present in the primary tumor. Deep sequencing of plasma ctDNA detected most DNA variants previously identified in matched primary and metastatic tissues. ctDNA might aid in therapy selection and in the search for biomarkers for drug development in metastatic breast cancer.

Neuronal and Glial Differentiation of Human Neural Stem Cells Is Regulated by Amyloid Precursor Protein (APP) Levels.

Amyloid precursor protein (APP) is implicated in neural development as well as in the pathology of Alzheimer's disease (AD); however, its biological function still remains unclear. It has been reported that APP stimulates the proliferation and neuronal differentiation of neural stem cells (NSCs), while other studies suggest an important effect enhancing gliogenesis in NSCs. As expected, APP protein/mRNA is detected in hNS1 cells, a model cell line of human NSCs, both under proliferation and throughout the differentiation period. To investigate the potential function that APP plays in cell fate specification and differentiation of hNS1 cells, we transiently increased human APP levels in these cells and analyzed its cell intrinsic effects. Our data indicate that increased levels of APP induce early cell cycle exit and instructively direct hNS1 cell fate towards a glial phenotype, while decreasing neuronal differentiation. Since elevated APP levels also enhanced APP intracellular domain (AICD)-immunoreactivity, these effects could be, in part, mediated by the APP/AICD system. The AICD domain can play a potential role in signal transduction by its molecular interaction with different target genes such as GSK3B, whose expression was also increased in APP-overexpressing cells that, in turn, may contribute to promoting gliogenesis and inhibiting neurogenesis in NSCs. These data suggest an important action of APP in modulating hNSCs differentiation (probably in an AICD-GSK-3ß-dependent manner) and may thus be important for the future development of stem cell therapy strategies for the diseased mammalian brain.

Pathological Response in a Triple-Negative Breast Cancer Cohort Treated with Neoadjuvant Carboplatin and Docetaxel According to Lehmann's Refined Classification.

Purpose: Triple-negative breast cancer (TNBC) requires the iden- tification of reliable predictors of response to neoadjuvant chemotherapy (NACT). For this purpose, we aimed to evaluate the performance of the TNBCtype-4 classifier in a cohort of patients with TNBC treated with neoadjuvant carboplatin and docetaxel (TCb).Methods: Patients with TNBC were accrued in a nonrandomized trial of neoadjuvant carboplatin AUC 6 and docetaxel 75 mg/m2 for six cycles. Response was evaluated in terms of pathologic complete response (pCR, ypT0/is ypN0) and residual cancer burden by Symmans and colleagues. Lehmann's subtyping was performed using the TNBCtype online tool from RNAseq data, and germline sequencing of a panel of seven DNA damage repair genes was conducted.Results: Ninety-four out of the 121 patients enrolled in the trial had RNAseq available. The overall pCR rate was 44.7%. Lehmann subtype distribution was 34.0% BL1, 20.2% BL2, 23.4% M, 14.9% LAR, and 7.4% were classified as ER+. Response to NACT with TCb was significantly associated with Lehmann subtype (P = 0.027), even in multivariate analysis including tumor size and nodal involvement, with BL1 patients achieving the highest pCR rate (65.6%), followed by BL2 (47.4%), M (36.4%), and LAR (21.4%). BL1 was associated with a significant younger age at diagnosis and higher ki67 values. Among our 10 germline mutation carriers, 30% were BL1, 40% were BL2, and 30% were M.Conclusions: TNBCtype-4 is associated with significantly different pCR rates for the different subtypes, with BL1 and LAR displaying the best and worse responses to NACT, respectively. Clin Cancer Res; 24(8); 1845-52. ©2018 AACR.

Frequency of germline DNA genetic findings in an unselected prospective cohort of triple-negative breast cancer patients participating in a platinum-based neoadjuvant chemotherapy trial.

We describe the status and frequency of germline DNA genetic findings in an unselected prospective cohort of triple negative breast cancer patients participating in a platinum-based neoadjuvant chemotherapy trial. Study population includes 124 consecutive patients with stage II-III TNBC from a trial exploring the antitumor activity of neoadjuvant carboplatin/docetaxel chemotherapy enrolled between 2012 and March 2015, to determine the frequency of germline DNA genetic mutations. 17.1 % of the patients with germline DNA tested had deleterious mutations in any of the analyzed genes (12.38 % in BRCA1, 1.9 % in BRCA2 and BARD1 and 0.95 % in RAD51D). Attending the intrinsic subtype, all the BRCA1/2 carriers tested had basal-like subtype. Among wild-type (WT) patients, 70.11 % had basal subtype, 16.09 % HER2 enriched, 1.15 % Luminal B, and 4.60 % Normal-like. Mean age at diagnosis was significantly lower in mutation-carriers compared with no carriers (43.72 vs 53.10, p = 0.004). 3 BRCA1/2 carriers were detected between 51 and 60 years, and only one deleterious mutation (BARD1) over 60 years. A positive familiar history of breast and ovarian cancer was more frequent in patients with deleterious mutations (39.39 vs 17.94 %, p = 0.043). Our study confirms the prevalence of BRCA1/2 mutations in TNBC patients. TNBC should therefore be considered by itself as a criterion for BRCA1/2 genetic testing. Determination of other breast cancer predisposition genes implicated in homologous recombination should also be discussed in this population. However, no definitive conclusions can be reached due to the low prevalence and the uncertain clinical impact of most of the genes included.

FGF signaling enhances a sonic hedgehog negative feedback loop at the initiation of spinal cord ventral patterning.

A prevalent developmental mechanism for the assignment of cell identities is the production of spatiotemporal concentration gradients of extracellular signaling molecules that are interpreted by the responding cells. One of such signaling systems is the Shh gradient that controls neuronal subtype identity in the ventral spinal cord. Using loss and gain of function approaches in chick and mouse embryos, we show here that the fibroblast growth factor (FGF) signaling pathway is required to restrict the domains of ventral gene expression as neuroepithelial cells become exposed to Shh during caudal extension of the embryo. FGF signaling activates the expression of the Shh receptor and negative pathway regulator Patched 2 (Ptch2) and therefore can enhance a negative feedback loop that restrains the activity of the pathway. Thus, we identify one of the mechanisms by which FGF signaling acts as a modulator of the onset of Shh signaling activity in the context of coordination of ventral patterning and caudal axis extension. © 2015 Wiley Periodicals, Inc. Develop Neurobiol 76: 956-971, 2016.

Appropriate Bmp7 levels are required for the differentiation of midline guidepost cells involved in corpus callosum formation.

Guidepost cells are essential structures for the establishment of major axonal tracts. How these structures are specified and acquire their axon guidance properties is still poorly understood. Here, we show that in mouse embryos appropriate levels of Bone Morphogenetic Protein 7 (Bmp7), a member of the TGF-ß superfamily of secreted proteins, are required for the correct development of the glial wedge, the indusium griseum, and the subcallosal sling, three groups of cells that act as guidepost cells for growing callosal axons. Bmp7 is expressed in the region occupied by these structures and its genetic inactivation in mouse embryos caused a marked reduction and disorganization of these cell populations. On the contrary, infusion of recombinant Bmp7 in the developing forebrain induced their premature differentiation. In both cases, changes were associated with the disruption of callosal axon growth and, in most animals fibers did not cross the midline forming typical Probst bundles. Addition of Bmp7 to cortical explants did not modify the extent of their outgrowth nor their directionality, when explants were exposed to a focalized source of the protein. Together, these results indicate that Bmp7 is indirectly required for corpus callosum formation by controlling the timely differentiation of its guidepost cells.