Publications by authors named "Hannah Gilmore"

57 Publications

Low-Grade Mucoepidermoid Carcinoma Versus Nodular Hidradenoma: Potential Diagnostic Challenge in Breast Pathology.

Int J Surg Pathol 2020 Dec 16:1066896920981635. Epub 2020 Dec 16.

University Hospitals Cleveland Medical Center, Cleveland, OH, USA.

View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1177/1066896920981635DOI Listing
December 2020

Magnetic Resonance Molecular Imaging of Extradomain B Fibronectin Improves Imaging of Pancreatic Cancer Tumor Xenografts.

Front Oncol 2020 30;10:586727. Epub 2020 Oct 30.

Department of Biomedical Engineering, Case School of Engineering, Case Western Reserve University, Cleveland, OH, United States.

The survival of pancreatic cancer patients can be greatly improved if their disease is detected at an early, potentially curable stage. Magnetic resonance molecular imaging (MRMI) of oncoproteins is a promising strategy for accurate, early detection of the disease. Here, we test the hypothesis that MRMI of extradomain-B fibronectin (EDB-FN), an abundant oncoprotein in the tumor extracellular matrix, can overcome the stromal barriers of pancreatic cancer to facilitate effective molecular imaging and detection of small tumors. Specimens of normal, premalignant, and malignant human pancreatic tissues were stained with a peptide-fluorophore conjugate (ZD2-Cy5.5) to assess EDB-FN binding and expression. MRMI with ZD2-N3-Gd(HP-DO3A) (MT218) specific to EDB-FN and MRI with Gd(HP-DO3A) were performed in three murine models bearing human pancreatic cancer xenografts, including a Capan-1 flank model, a BxPC3-GFP-Luc and a PANC-1-GFP-Luc intrapancreatic xenograft model. Tumor enhancement of the contrast agents was analyzed and compared. Staining of human tissue samples with ZD2-Cy5.5 revealed high EDB-FN expression in pancreatic tumors, moderate expression in premalignant tissue, and little expression in normal tissue. MRMI with MT218 generated robust intratumoral contrast, clearly detected and delineated small tumors (smallest average size: 6.1 mm), and out-performed conventional contrast enhanced MRI with Gd(HP-DO3A). Quantitative analysis of signal enhancement revealed that MT218 produced 2.7, 2.1, and 1.6 times greater contrast-to-noise ratio (CNR) than the clinical agent in the Capan-1 flank, BxPC3-GFP-Luc and PANC-1-GFP-Luc intrapancreatic models, respectively ( < 0.05). MRMI of the ECM oncoprotein EDB-FN with MT218 is able to generate superior contrast enhancement in small pancreatic tumors and provide accurate tumor delineation in animal models. Early, accurate detection and delineation of pancreatic cancer with high-resolution MRMI has the potential to guide timely treatment and significantly improve the long-term survival of pancreatic cancer patients.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.3389/fonc.2020.586727DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC7661967PMC
October 2020

Noninvasive assessment and therapeutic monitoring of drug-resistant colorectal cancer by MR molecular imaging of extradomain-B fibronectin.

Theranostics 2020 8;10(24):11127-11143. Epub 2020 Sep 8.

Department of Biomedical Engineering, Case Western Reserve University, Cleveland, OH 44106, USA.

Antineoplastic resistance represents a multifaceted challenge for cancer therapy and diagnostics. Extensive molecular heterogeneity, even within neoplasms of the same type, can elicit distinct outcomes of administering therapeutic pressures, frequently leading to the development of drug-resistant populations. Improved success of oncotherapies merits the exploration of precise molecular imaging technologies that can detect not only anatomical but also molecular changes in tumors and their microenvironment, early on in the treatment regimen. To this end, we developed magnetic resonance molecular imaging (MRMI) strategies to target the extracellular matrix oncoprotein, extradomain-B fibronectin (EDB-FN), for non-invasive assessment and therapeutic monitoring of drug-resistant colorectal cancer (CRC). Two drug-resistant CRC lines generated from parent DLD-1 and RKO cells by long-term treatment with 5'-FU and 5'-FU plus CB-839 respectively, were characterized for functional and gene expression changes using 3D culture, transwell invasion, qRT-PCR, and western blot assays. Contrast-enhanced MRMI of EDB-FN was performed in athymic nu/nu mice bearing subcutaneous tumor xenografts with 40 µmol/kg dose of macrocyclic ZD2-targeted contrast agent MT218 [ZD2-N-Gd (HP-DO3A)] on a 3T MRS 3000 scanner. Immunohistochemistry was conducted on patient specimens and xenografts using anti-EDB-FN antibody G4. Analyses of TCGA and GTEx databases revealed poor prognosis of colon cancer patients with higher levels of EDB-FN. Similarly, immunohistochemical staining of patient specimens showed increased EDB-FN expression in primary colon adenocarcinoma and hepatic metastases, but none in normal adjacent tissues. Drug-resistant DLD1-DR and RKO-DR cells were also found to demonstrate enhanced invasive potential and significantly elevated EDB-FN expression over their parent counterparts. MRMI of EDB-FN with 40 µmol/kg dose of MT218 (60% lower than the clinical dose) resulted in robust signal enhancement in the drug-resistant CRC xenografts with 84-120% increase in their contrast-to-noise ratios (CNRs) over the non-resistant counterparts. The feasibility of non-invasive therapeutic monitoring using MRMI of EDB-FN was also evaluated in drug-resistant DLD1-DR tumors treated with a pan-AKT inhibitor MK2206-HCl. The treated drug-resistant tumors failed to respond to therapy, which was accurately detected by MRMI with MT218, demonstrating higher signal enhancement and increased CNRs in the 4-week follow-up scans over the pre-treatment scans. EDB-FN is a promising molecular marker for assessing drug resistance. MRMI of EDB-FN with MT218 at a significantly reduced dose can facilitate effective non-invasive assessment and treatment response monitoring of drug-resistant CRC, highlighting its translational potential for active surveillance and management of CRC and other malignancies.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.7150/thno.47448DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC7532678PMC
September 2020

Correlation between modified Magee equation-2 and Oncotype-Dx recurrence scores using both traditional and TAILORx cutoffs and the clinical application of the Magee Decision Algorithm: a single institutional review.

Breast Cancer 2021 Mar 20;28(2):321-328. Epub 2020 Sep 20.

Department of Pathology, University Hospitals Cleveland Medical Center/Case Western Reserve University, Cleveland, Ohio, USA.

Background: Oncotype Dx (ODX) is used to predict recurrence risk for estrogen-positive (ER +), HER2-negative and lymph node negative breast cancer, however, due to the cost its use may be limited in low-resource areas. The aim of this study is to assess the concordance between the modified Magee Equation-2 (MME-2) and ODX recurrence scores (RS). The secondary aim is to apply the Magee Decision Algorithm (MDA) using the MME-2 to determine which patients are unlikely to benefit from ODX testing.

Methods: All newly diagnosed ER + , HER2 negative, lymph node negative breast cancer patients with available ODX-RS from 2008-2018 were included. The original pathology reports were reviewed and chart review was performed. The MME-2 scores were calculated and correlated with the ODX-RS. The MDA was applied to our cohort to assess which patients would not benefit from ODX testing.

Results: A total of 579 patients were included. There was an overall moderate correlation between ODX-RS and MME-2 score (Pearson correlation coefficient = 0.635). The overall concordance between ODX and MME-2 scores was similar when using both the traditional and TAILORx cutoffs (63.3% vs. 63.7%, respectively). Applying the MDA, for patients with MME-2 scores < 18, 96.8% of patients had the expected ODX-RS of < 25. For patients with MME-2 RS > 30, 90% had the expected ODX-RS of > 25. Concordance was highest in the high-risk category using both cutoffs. For patients with MME-2 18-25 and a mitotic score of 1, 88.8% had the expected ODX-RS of > 25.

Conclusion: There is a moderate correlation between MME-2 score and ODX-RS. The overall concordance was similar for both traditional and TAILORx cutoffs. The strongest concordance was found in the high-risk category for both cutoffs. The MME-2 can be used to identify patients unlikely to benefit from ODX testing using the MDA.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1007/s12282-020-01163-3DOI Listing
March 2021

Clinicopathological and radiological characterization of myofibroblastoma of breast: A single institutional case review.

Ann Diagn Pathol 2020 Oct 15;48:151591. Epub 2020 Aug 15.

Department of Pathology, University Hospitals Cleveland Medical Center, Cleveland, OH, USA; Case Western Reserve University, Cleveland, OH, USA.

Myofibroblastoma is a rare type of benign myofibroblastic neoplasm in the breast. It is clinically presented as a well-circumscribed mass, usually small in size (usually less than 4.0 cm), and can mostly be cured by local excision. Rare cases of giant myofibroblastoma greater than 10 cm have been reported, but also follow a benign clinical course. Histologically, breast myofibroblastoma is featured by bland fascicles of spindle cells intermixed with thick hyalinized collagen bundles. Mast cells are frequently found within the stroma. However, a wide spectrum of morphological variants can occur in myofibroblatoma, making its diagnosis challenging sometimes. Differential diagnosis of myofibroblastoma with other spindle cell lesions in the breast, either benign or malignant, is also important in practice. In this study, we collected 15 cases of breast myofibroblastoma diagnosed in our institution during a 20 year period. The sizes of these cases range from 0.4 cm to 35.2 cm (mean is 3.7 cm). To our knowledge, the case of giant breast myofibroblastoma we presented here is the largest one reported to date. The histological examination of the cases show great morphological variations. Besides the classical type, features of cellular, collagenized, palisading, epithelioid, myxoid, myoid, solitary fibrous tumor-like are also identified in the case series. Immunohistochemical staining patterns as well as clinical features of the cases are also summarized and compared. All cases in this study show no recurrence on follow-up. In addition, cases that are important differential diagnosis for breast myofibroblastoma are also studied. Their key histological characteristics are compared with myofibroblastoma, and their immunohistochemical and molecular features are discussed.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.anndiagpath.2020.151591DOI Listing
October 2020

Concordance of breast cancer biomarker status between routine immunohistochemistry/in situ hybridization and Oncotype DX qRT-PCR with investigation of discordance, a study of 591 cases.

Hum Pathol 2020 10 3;104:54-65. Epub 2020 Aug 3.

Department of Pathology, University Hospitals Cleveland Medical Center, Cleveland, OH, 44106, USA.

Patients with estrogen receptor (ER)+/human epidermal growth factor receptor (HER)2-, lymph node- breast cancer with high recurrence risk benefit from adjuvant chemotherapy in addition to hormonal therapy. This study compares ER, progesterone receptor (PR), and HER2 status between routine immunohistochemistry (IHC)/in situ hybridization (ISH) and Oncotype DX (ODX) in 591 cases. ODX recurrence score (RS) and clinicopathologic features were compared between ER/PR-concordant and discordant cases. Hematoxylin and eosin (H&E) slides from ER discordant cases were reexamined. Concordance was high between ODX and IHC for ER status (580/591, 98.1%) and moderate for PR status (512/591, 86.6%). All 11 ER discordant cases were ER+ by IHC but ER- by ODX and high risk by ODX. Histologically, all of these cases were grade III invasive ductal carcinoma (IDC), except one case diagnosed as IDC with apocrine features. Although this case was grade I and ER/PR+ by IHC, this patient received chemotherapy because of high RS. Of 79 PR discordant cases, 60 were PR+ by IHC but PR- by ODX. Five hundred eighty-four cases had available HER2 data, with high negative agreement (580/582, 99.7%). However, both HER2+ cases by ISH were HER2- by ODX. Mean RS was higher for ER discordant than concordant cases (48.0 versus 17.1, P < 0.0001) and for PR discordant (IHC+/ODX-) than concordant cases (27.2 versus 16.7, P < 0.0001) with no significant differences in recurrence or metastasis. Overall, detection was more sensitive by IHC, and high RS of discordant cases suggests possible risk overestimation. Therapeutic decisions for discordant cases should continue to be based on clinicopathologic correlation and not oncotype alone.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.humpath.2020.07.022DOI Listing
October 2020

Overexpression of Extradomain-B Fibronectin is Associated with Invasion of Breast Cancer Cells.

Cells 2020 08 3;9(8). Epub 2020 Aug 3.

Department of Biomedical Engineering, Case Western Reserve University, Cleveland, OH 44106, USA.

Breast tumor heterogeneity is a major impediment to oncotherapy. Cancer cells undergo rapid clonal evolution, thereby acquiring significant growth and invasive advantages. The absence of specific markers of these high-risk populations precludes efficient therapeutic and diagnostic management of the disease. Given the critical function of tumor microenvironment in the oncogenic circuitry, we sought to determine the expression profile of the extracellular matrix oncoprotein, extradomain-B fibronectin (EDB-FN) in invasive breast cancer. Analyses of TCGA/GTEx databases and immunostaining of clinical samples found a significant overexpression of EDB-FN in breast tumors, which correlated with poor overall survival. Significant upregulation of EDB-FN was observed in invasive cell populations generated from relatively less invasive MCF7 and MDA-MB-468 cells by long-term TGF-β treatment and acquired chemoresistance. Treatment of the invasive cell populations with an AKT inhibitor (MK2206-HCl) reduced their invasive potential, with a concomitant decrease in their EDB-FN expression, partly through the phosphoAKT-SRp55 pathway. EDB-FN downregulation, with direct RNAi of EDB-FN or indirectly through RNAi of SRp55, also resulted in reduced motility of the invasive cell populations, validating the correlation between EDB-FN expression and invasion of breast cancer cells. These data establish EDB-FN as a promising molecular marker for non-invasive therapeutic surveillance of aggressive breast cancer.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.3390/cells9081826DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC7463489PMC
August 2020

Molecular Detection of SARS-CoV-2 Infection in FFPE Samples and Histopathologic Findings in Fatal SARS-CoV-2 Cases.

Am J Clin Pathol 2020 07;154(2):190-200

Department of Pathology, University Hospitals Cleveland Medical Center, Cleveland, OH.

Objectives: To report methods and findings of 2 autopsies with molecular evaluation of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) positive individuals.

Methods: Postmortem examination was completed following Centers for Disease Control and Prevention public guidelines. Numerous formalin-fixed paraffin-embedded (FFPE) tissue types from each case were surveyed for SARS-CoV-2 RNA by quantitative reverse transcription polymerase chain reaction (qRT-PCR). SARS-CoV-2 viral genome was sequenced by next-generation sequencing (NGS) from FFPE lung tissue blocks.

Results: Postmortem examinations revealed diffuse alveolar damage, while no viral-associated hepatic, cardiac, or renal damage was observed. Viral RNA was detected in lungs, bronchi, lymph nodes, and spleen in both cases using qRT-PCR method. RNA sequencing using NGS in case 1 revealed mutations most consistent with Western European Clade A2a with ORF1a L3606F mutation.

Conclusions: SARS-CoV-2 testing and viral sequencing can be performed from FFPE tissue. Detection and sequencing of SARS-CoV-2 in combination with morphological findings from postmortem tissue examination can aid in gaining a better understanding of the virus's pathophysiologic effects on human health.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1093/ajcp/aqaa091DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC7314275PMC
July 2020

Mucoepidermoid Carcinoma of the Breast With Rearrangement: A Case Report and Literature Review.

Int J Surg Pathol 2020 Oct 4;28(7):787-792. Epub 2020 May 4.

University Hospitals Cleveland Medical Center, Cleveland, OH, USA.

Mucoepidermoid carcinoma is one of the most common malignancies in salivary glands. In comparison, breast mucoepidermoid carcinoma is a very rare entity, with limited reports and understanding of its clinical behaviors to date. In this article, we report a case of low-grade breast mucoepidermoid carcinoma of a 60-year-old female patient. Histologic and immunohistochemical patterns were demonstrated. Fluorescence in situ hybridization test was also conducted to identify rearrangement in this case, indicating a similar molecular abnormality as mucoepidermoid carcinoma in the salivary gland. Five-year follow-up of the patient showed no local recurrence or distant metastasis of the carcinoma, indicating the indolent behavior of low-grade breast mucoepidermoid carcinoma. Besides, a 40-year literature review from 1979 to 2019 was also performed to better characterize the prognosis and molecular abnormalities of the lesion.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1177/1066896920916779DOI Listing
October 2020

Quantitative nuclear histomorphometric features are predictive of Oncotype DX risk categories in ductal carcinoma in situ: preliminary findings.

Breast Cancer Res 2019 10 17;21(1):114. Epub 2019 Oct 17.

Department of Biomedical Engineering, Case Western Reserve University, Cleveland, OH, USA.

Background: Oncotype DX (ODx) is a 12-gene assay assessing the recurrence risk (high, intermediate, and low) of ductal carcinoma in situ (pre-invasive breast cancer), which guides clinicians regarding prescription of radiotherapy. However, ODx is expensive, time-consuming, and tissue-destructive. In addition, the actual prognostic meaning for the intermediate ODx risk category remains unclear.

Methods: In this work, we evaluated the ability of quantitative nuclear histomorphometric features extracted from hematoxylin and eosin-stained slide images of 62 ductal carcinoma in situ (DCIS) patients to distinguish between the corresponding ODx risk categories. The prognostic value of the identified image signature was further evaluated on an independent validation set of 30 DCIS patients in its ability to distinguish those DCIS patients who progressed to invasive carcinoma versus those who did not. Following nuclear segmentation and feature extraction, feature ranking strategies were employed to identify the most discriminating features between individual ODx risk categories. The selected features were then combined with machine learning classifiers to establish models to predict ODx risk categories. The model performance was evaluated using the average area under the receiver operating characteristic curve (AUC) using cross validation. In addition, an unsupervised clustering approach was also implemented to evaluate the ability of nuclear histomorphometric features to discriminate between the ODx risk categories.

Results: Features relating to spatial distribution, orientation disorder, and texture of nuclei were identified as most discriminating between the high ODx and the intermediate, low ODx risk categories. Additionally, the AUC of the most discriminating set of features for the different classification tasks was as follows: (1) high vs low ODx (0.68), (2) high vs. intermediate ODx (0.67), (3) intermediate vs. low ODx (0.57), (4) high and intermediate vs. low ODx (0.63), (5) high vs. low and intermediate ODx (0.66). Additionally, the unsupervised clustering resulted in intermediate ODx risk category patients being co-clustered with low ODx patients compared to high ODx.

Conclusion: Our results appear to suggest that nuclear histomorphometric features can distinguish high from low and intermediate ODx risk category patients. Additionally, our findings suggest that histomorphometric features for intermediate ODx were more similar to low ODx compared to high ODx risk category.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1186/s13058-019-1200-6DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC6798488PMC
October 2019

Enlarging biopsy-proven fibroadenoma: Is surgical excision necessary?

Clin Imaging 2019 Sep - Oct;57:35-39. Epub 2019 Apr 4.

Beth Israel Deaconess Medical Center, Department of Radiology, 330 Brookline Avenue, Boston, MA 02215, United States of America.

Purpose: Core biopsy-proven fibroadenomas that enlarge on clinical or imaging follow-up are often surgically excised to exclude an associated malignancy. The purpose of this study was to assess how often malignancy is detected upon excision, and to determine whether excision of enlarging, biopsy-proven, uncomplicated fibroadenomas is still warranted.

Materials And Methods: Review of our institutional pathology database from 2000 to 2010 identified 1117 cases of fibroadenoma, and retrospective chart review, including review of pathology and imaging findings of all these records, was performed.

Results: 1117 cases of fibroadenoma were identified in a population of women ranging from ages 17 to 78. Of these, 378 (33.8%) were diagnosed by ultrasound core needle biopsy and formed the study population. Of the 378 cases, 24 (6.3%) had co-existent atypia and were immediately excised; these cases were excluded. An additional 107 (28%) were lost to follow-up. Of the remaining 247 cases, 201 (81%) showed stability on follow-up imaging (mean 31.5 months), and 46 (18.6%) enlarged on follow-up. Of the 46 biopsy proven fibroadenomas that enlarged, 19 had a biopsy at initial presentation and 27 underwent biopsy after they enlarged. Seventeen of the 19 were excised after enlargement, and pathology confirmed fibroadenoma in all cases (100%); two enlarged on initial follow-up imaging but remained stable for at least three years on continued follow-up. Of the 27 cases which were biopsied after enlargement, 23 revealed fibroadenoma on core biopsy, 3 had fibroadenoma with associated atypia with subsequent surgery revealing fibroadenoma and no associated malignancy, and one showed fibroadenoma with smooth muscle with subsequent surgery showing phyllodes tumor.

Conclusion: Based on this study, enlarging biopsy proven fibroadenomas are not associated with malignancy; therefore, surgical excision does not seem warranted. For presumed enlarging fibroadenomas on imaging, core biopsy should be performed to exclude associated atypia or phyllodes tumor. Finally, surgical excision is indicated for lesions with associated atypia or suspected phyllodes and for symptomatic lesions or cosmetic reasons.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.clinimag.2019.03.014DOI Listing
January 2020

Association of Peritumoral Radiomics With Tumor Biology and Pathologic Response to Preoperative Targeted Therapy for HER2 (ERBB2)-Positive Breast Cancer.

JAMA Netw Open 2019 04 5;2(4):e192561. Epub 2019 Apr 5.

Department of Biomedical Engineering, Case Western Reserve University, Cleveland, Ohio.

Importance: There has been significant recent interest in understanding the utility of quantitative imaging to delineate breast cancer intrinsic biological factors and therapeutic response. No clinically accepted biomarkers are as yet available for estimation of response to human epidermal growth factor receptor 2 (currently known as ERBB2, but referred to as HER2 in this study)-targeted therapy in breast cancer.

Objective: To determine whether imaging signatures on clinical breast magnetic resonance imaging (MRI) could noninvasively characterize HER2-positive tumor biological factors and estimate response to HER2-targeted neoadjuvant therapy.

Design, Setting, And Participants: In a retrospective diagnostic study encompassing 209 patients with breast cancer, textural imaging features extracted within the tumor and annular peritumoral tissue regions on MRI were examined as a means to identify increasingly granular breast cancer subgroups relevant to therapeutic approach and response. First, among a cohort of 117 patients who received an MRI prior to neoadjuvant chemotherapy (NAC) at a single institution from April 27, 2012, through September 4, 2015, imaging features that distinguished HER2+ tumors from other receptor subtypes were identified. Next, among a cohort of 42 patients with HER2+ breast cancers with available MRI and RNaseq data accumulated from a multicenter, preoperative clinical trial (BrUOG 211B), a signature of the response-associated HER2-enriched (HER2-E) molecular subtype within HER2+ tumors (n = 42) was identified. The association of this signature with pathologic complete response was explored in 2 patient cohorts from different institutions, where all patients received HER2-targeted NAC (n = 28, n = 50). Finally, the association between significant peritumoral features and lymphocyte distribution was explored in patients within the BrUOG 211B trial who had corresponding biopsy hematoxylin-eosin-stained slide images. Data analysis was conducted from January 15, 2017, to February 14, 2019.

Main Outcomes And Measures: Evaluation of imaging signatures by the area under the receiver operating characteristic curve (AUC) in identifying HER2+ molecular subtypes and distinguishing pathologic complete response (ypT0/is) to NAC with HER2-targeting.

Results: In the 209 patients included (mean [SD] age, 51.1 [11.7] years), features from the peritumoral regions better discriminated HER2-E tumors (maximum AUC, 0.85; 95% CI, 0.79-0.90; 9-12 mm from the tumor) compared with intratumoral features (AUC, 0.76; 95% CI, 0.69-0.84). A classifier combining peritumoral and intratumoral features identified the HER2-E subtype (AUC, 0.89; 95% CI, 0.84-0.93) and was significantly associated with response to HER2-targeted therapy in both validation cohorts (AUC, 0.80; 95% CI, 0.61-0.98 and AUC, 0.69; 95% CI, 0.53-0.84). Features from the 0- to 3-mm peritumoral region were significantly associated with the density of tumor-infiltrating lymphocytes (R2 = 0.57; 95% CI, 0.39-0.75; P = .002).

Conclusions And Relevance: A combination of peritumoral and intratumoral characteristics appears to identify intrinsic molecular subtypes of HER2+ breast cancers from imaging, offering insights into immune response within the peritumoral environment and suggesting potential benefit for treatment guidance.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1001/jamanetworkopen.2019.2561DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC6481453PMC
April 2019

HistoQC: An Open-Source Quality Control Tool for Digital Pathology Slides.

JCO Clin Cancer Inform 2019 04;3:1-7

Case Western Reserve University, Cleveland, OH.

Purpose: Digital pathology (DP), referring to the digitization of tissue slides, is beginning to change the landscape of clinical diagnostic workflows and has engendered active research within the area of computational pathology. One of the challenges in DP is the presence of artefacts and batch effects, unintentionally introduced during both routine slide preparation (eg, staining, tissue folding) and digitization (eg, blurriness, variations in contrast and hue). Manual review of glass and digital slides is laborious, qualitative, and subject to intra- and inter-reader variability. Therefore, there is a critical need for a reproducible automated approach of precisely localizing artefacts to identify slides that need to be reproduced or regions that should be avoided during computational analysis.

Methods: Here we present HistoQC, a tool for rapidly performing quality control to not only identify and delineate artefacts but also discover cohort-level outliers (eg, slides stained darker or lighter than others in the cohort). This open-source tool employs a combination of image metrics (eg, color histograms, brightness, contrast), features (eg, edge detectors), and supervised classifiers (eg, pen detection) to identify artefact-free regions on digitized slides. These regions and metrics are presented to the user via an interactive graphical user interface, facilitating artefact detection through real-time visualization and filtering. These same metrics afford users the opportunity to explicitly define acceptable tolerances for their workflows.

Results: The output of HistoQC on 450 slides from The Cancer Genome Atlas was reviewed by two pathologists and found to be suitable for computational analysis more than 95% of the time.

Conclusion: These results suggest that HistoQC could provide an automated, quantifiable, quality control process for identifying artefacts and measuring slide quality, in turn helping to improve both the repeatability and robustness of DP workflows.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1200/CCI.18.00157DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC6552675PMC
April 2019

Convolutional neural network initialized active contour model with adaptive ellipse fitting for nuclear segmentation on breast histopathological images.

J Med Imaging (Bellingham) 2019 Jan 8;6(1):017501. Epub 2019 Feb 8.

Case Western Reserve University, Department of Biomedical Engineering, Cleveland, Ohio, United States.

Automated detection and segmentation of nuclei from high-resolution histopathological images is a challenging problem owing to the size and complexity of digitized histopathologic images. In the context of breast cancer, the modified Bloom-Richardson Grading system is highly correlated with the morphological and topological nuclear features are highly correlated with Modified Bloom-Richardson grading. Therefore, to develop a computer-aided prognosis system, automated detection and segmentation of nuclei are critical prerequisite steps. We present a method for automated detection and segmentation of breast cancer nuclei named a convolutional neural network initialized active contour model with adaptive ellipse fitting (CoNNACaeF). The CoNNACaeF model is able to detect and segment nuclei simultaneously, which consist of three different modules: convolutional neural network (CNN) for accurate nuclei detection, (2) region-based active contour (RAC) model for subsequent nuclear segmentation based on the initial CNN-based detection of nuclear patches, and (3) adaptive ellipse fitting for overlapping solution of clumped nuclear regions. The performance of the CoNNACaeF model is evaluated on three different breast histological data sets, comprising a total of 257 H&E-stained images. The model is shown to have improved detection accuracy of F-measure 80.18%, 85.71%, and 80.36% and average area under precision-recall curves (AveP) 77%, 82%, and 74% on a total of 3 million nuclei from 204 whole slide images from three different datasets. Additionally, CoNNACaeF yielded an F-measure at 74.01% and 85.36%, respectively, for two different breast cancer datasets. The CoNNACaeF model also outperformed the three other state-of-the-art nuclear detection and segmentation approaches, which are blue ratio initialized local region active contour, iterative radial voting initialized local region active contour, and maximally stable extremal region initialized local region active contour models.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1117/1.JMI.6.1.017501DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC6368488PMC
January 2019

Systemic Delivery of Tumor-Targeting siRNA Nanoparticles against an Oncogenic LncRNA Facilitates Effective Triple-Negative Breast Cancer Therapy.

Bioconjug Chem 2019 03 21;30(3):907-919. Epub 2019 Feb 21.

University Hospitals of Cleveland , Department of Pathology , Cleveland , Ohio 44106 , United States.

Long noncoding RNAs (lncRNAs), by virtue of their versatility and multilevel gene regulation, have emerged as attractive pharmacological targets for treating heterogeneous and complex malignancies like triple-negative breast cancer (TNBC). Despite multiple studies on lncRNA functions in tumor pathology, systemic targeting of these "undruggable" macromolecules with conventional approaches remains a challenge. Here, we demonstrate effective TNBC therapy by nanoparticle-mediated RNAi of the oncogenic lncRNA DANCR, which is significantly overexpressed in TNBC. Tumor-targeting RGD-PEG-ECO/siDANCR nanoparticles were formulated via self-assembly of multifunctional amino lipid ECO, cyclic RGD peptide-PEG, and siDANCR for systemic delivery. MDA-MB-231 and BT549 cells treated with the therapeutic RGD-PEG-ECO/siDANCR nanoparticles exhibited 80-90% knockdown in the expression of DANCR for up to 7 days, indicating efficient intracellular siRNA delivery and sustained target silencing. The RGD-PEG-ECO/siDANCR nanoparticles mediated excellent in vitro therapeutic efficacy, reflected by significant reduction in the invasion, migration, survival, tumor spheroid formation, and proliferation of the TNBC cell lines. At the molecular level, functional ablation of DANCR dynamically impacted the oncogenic nexus by downregulating PRC2-mediated H3K27-trimethylation and Wnt/EMT signaling, and altering the phosphorylation profiles of several kinases in the TNBC cells. Furthermore, systemic administration of the RGD-PEG-ECO/siDANCR nanoparticles at a dose of 1 mg/kg siRNA in nude mice bearing TNBC xenografts resulted in robust suppression of TNBC progression with no overt toxic side-effects, underscoring the efficacy and safety of the nanoparticle therapy. These results demonstrate that nanoparticle-mediated modulation of onco-lncRNAs and their molecular targets is a promising approach for developing curative therapies for TNBC and other cancers.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1021/acs.bioconjchem.9b00028DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC6820053PMC
March 2019

Clinical, imaging, and intervention factors associated with the upgrade of isolated flat epithelial atypia.

Clin Imaging 2019 Mar - Apr;54:21-24. Epub 2018 Nov 20.

Mather Pavilion B402, 11100 Euclid Avenue, Cleveland, OH, 44106, United States of America. Electronic address:

Purpose: This study aims to determine clinical, imaging, and intervention factors associated with the upgrade of flat epithelial atypia (FEA) diagnosed on vacuum-assisted biopsy (VAB) in order to formulate criteria for excision and better assist in management.

Methods: Between 2012 and 2015, 254 patients had a form of atypia diagnosed on ultrasound, MRI or stereotactic VAB and met eligibility for this study. Demographic, imaging, biopsy and pathology characteristics were analyzed for association with upgrade. We compared isolated FEA to all of the atypias grouped together.

Results: Of the 254 atypia lesions, 72 (28%) were isolated FEA, and the upgrade rate was 2.8% (2/72). Statistically significant factors present with upgrade of isolated FEA include personal history of breast cancer and cancer diagnosis on a concurrent separate core biopsy. Other factors associated with upgrade include first degree family history of breast cancer, segmental calcification distribution, extent of calcifications >2 cm, and <25% of calcifications removed on biopsy.

Conclusion: In patients with biopsy results of isolated FEA, in the absence of personal or first degree family history of breast cancer, cancer on a concurrent biopsy, segmental calcification distribution, extent of calcifications >2 cm, and only 0-24% calcifications removed on biopsy, patients may be safely followed with imaging, avoiding unnecessary excision.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.clinimag.2018.11.008DOI Listing
May 2019

Homophilic CD44 Interactions Mediate Tumor Cell Aggregation and Polyclonal Metastasis in Patient-Derived Breast Cancer Models.

Cancer Discov 2019 01 25;9(1):96-113. Epub 2018 Oct 25.

Department of Pharmacology, Feinberg School of Medicine, Northwestern University, Chicago, Illinois.

Circulating tumor cells (CTC) seed cancer metastases; however, the underlying cellular and molecular mechanisms remain unclear. CTC clusters were less frequently detected but more metastatic than single CTCs of patients with triple-negative breast cancer and representative patient-derived xenograft models. Using intravital multiphoton microscopic imaging, we found that clustered tumor cells in migration and circulation resulted from aggregation of individual tumor cells rather than collective migration and cohesive shedding. Aggregated tumor cells exhibited enriched expression of the breast cancer stem cell marker CD44 and promoted tumorigenesis and polyclonal metastasis. Depletion of CD44 effectively prevented tumor cell aggregation and decreased PAK2 levels. The intercellular CD44-CD44 homophilic interactions directed multicellular aggregation, requiring its N-terminal domain, and initiated CD44-PAK2 interactions for further activation of FAK signaling. Our studies highlight that CD44 CTC clusters, whose presence is correlated with a poor prognosis of patients with breast cancer, can serve as novel therapeutic targets of polyclonal metastasis. SIGNIFICANCE: CTCs not only serve as important biomarkers for liquid biopsies, but also mediate devastating metastases. CD44 homophilic interactions and subsequent CD44-PAK2 interactions mediate tumor cluster aggregation. This will lead to innovative biomarker applications to predict prognosis, facilitate development of new targeting strategies to block polyclonal metastasis, and improve clinical outcomes...
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1158/2159-8290.CD-18-0065DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC6328322PMC
January 2019

Nuclear shape and orientation features from H&E images predict survival in early-stage estrogen receptor-positive breast cancers.

Lab Invest 2018 11 29;98(11):1438-1448. Epub 2018 Jun 29.

Department of Biomedical Engineering, Case Western Reserve University, Cleveland, OH, USA.

Early-stage estrogen receptor-positive (ER+) breast cancer (BCa) is the most common type of BCa in the United States. One critical question with these tumors is identifying which patients will receive added benefit from adjuvant chemotherapy. Nuclear pleomorphism (variance in nuclear shape and morphology) is an important constituent of breast grading schemes, and in ER+ cases, the grade is highly correlated with disease outcome. This study aimed to investigate whether quantitative computer-extracted image features of nuclear shape and orientation on digitized images of hematoxylin-stained and eosin-stained tissue of lymph node-negative (LN-), ER+ BCa could help stratify patients into discrete (<10 years short-term vs. >10 years long-term survival) outcome groups independent of standard clinical and pathological parameters. We considered a tissue microarray (TMA) cohort of 276 ER+, LN- patients comprising 150 patients with long-term and 126 patients with short-term overall survival, wherein 177 randomly chosen cases formed the modeling set, and 99 remaining cases the test set. Segmentation of individual nuclei was performed using multiresolution watershed; subsequently, 615 features relating to nuclear shape/texture and orientation disorder were extracted from each TMA spot. The Wilcoxon's rank-sum test identified the 15 most prognostic quantitative histomorphometric features within the modeling set. These features were then subsequently combined via a linear discriminant analysis classifier and evaluated on the test set to assign a probability of long-term vs. short-term disease-specific survival. In univariate survival analysis, patients identified by the image classifier as high risk had significantly poorer survival outcome: hazard ratio (95% confident interval) = 2.91(1.23-6.92), p = 0.02786. Multivariate analysis controlling for T-stage, histology grade, and nuclear grade showed the classifier to be independently predictive of poorer survival: hazard ratio (95% confident interval) = 3.17(0.33-30.46), p = 0.01039. Our results suggest that quantitative histomorphometric features of nuclear shape and orientation are strongly and independently predictive of patient survival in ER+, LN- BCa.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1038/s41374-018-0095-7DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC6214731PMC
November 2018

Clinicopathological, immunohistochemical and molecular correlation of neural crest transcription factor SOX10 expression in triple-negative breast carcinoma.

Hum Pathol 2018 10 9;80:163-169. Epub 2018 Jun 9.

Department of Pathology, University Hospitals Cleveland Medical Center, 44106 Cleveland, OH.

The transcription factor SOX10 mediates the differentiation of neural crest-derived cells, and SOX10 by immunohistochemistry (IHC) is used primarily for the diagnosis of melanoma. SOX10 expression has been previously documented in benign breast myoepithelial cells. However there is limited literature on its expression in triple-negative breast carcinoma (TNBC). The aim was to study the clinical, pathologic and molecular profiles of SOX10+ tumors in TNBC. Tissue microarrays of TNBC were evaluated for SOX10 expression in 48 cases. SOX10 expression was correlated with clinical and pathologic features such as age, grade, and stage. Gene expression was analyzed on RNA extracted from formalin-fixed paraffin-embedded (FFPE) specimens with Affymetrix 2.0 HTA. Co-expression of SOX10 with androgen receptor (AR), WT1, gross cystic disease fluid protein-15 (GCDFP-15), mammaglobin, epidermal growth factor receptor (EGFR), CK5/6 and GATA transcription factor 3 (GATA3) were also assessed. The mean age was 59.38 (range, 28-90 years). Overall, 37.5% cases (18/48) were SOX10+. There was no association between SOX10 expression and age, grade or stage of patients; 6 of 10 (60%) cases of basal-like 1 (BL1), and 5 of 8 cases of unstable (UNS) molecular subtype were SOX10+. One of 5 basal-like-2 (BL2), 1 of 6 immunomodulatory (IM), 1 of 4 mesenchymal (M), 1 of 5 luminal androgen receptor (LAR) and 2 of 8 mesenchymal stem cell (MSL) showed lower frequencies of SOX10 expression. There was negative correlation between SOX10 and AR+ subtypes (P < .002). SOX10 was positively correlated with WT1 (P = .05). SOX10 did not show significant correlation with mammaglobin, GCDFP15, EGFR, CK5/6 and GATA3. SOX10 expression in the basal-like and unstable molecular subtypes supports the concept that these neoplasms show myoepithelial differentiation.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.humpath.2018.06.007DOI Listing
October 2018

Quantitative nuclear histomorphometry predicts oncotype DX risk categories for early stage ER+ breast cancer.

BMC Cancer 2018 May 30;18(1):610. Epub 2018 May 30.

Department of Biomedical Engineering, Case Western Reserve University, 2071 Martin Luther King Drive, Cleveland, OH, 44106-7207, USA.

Background: Gene-expression companion diagnostic tests, such as the Oncotype DX test, assess the risk of early stage Estrogen receptor (ER) positive (+) breast cancers, and guide clinicians in the decision of whether or not to use chemotherapy. However, these tests are typically expensive, time consuming, and tissue-destructive.

Methods: In this paper, we evaluate the ability of computer-extracted nuclear morphology features from routine hematoxylin and eosin (H&E) stained images of 178 early stage ER+ breast cancer patients to predict corresponding risk categories derived using the Oncotype DX test. A total of 216 features corresponding to the nuclear shape and architecture categories from each of the pathologic images were extracted and four feature selection schemes: Ranksum, Principal Component Analysis with Variable Importance on Projection (PCA-VIP), Maximum-Relevance, Minimum Redundancy Mutual Information Difference (MRMR MID), and Maximum-Relevance, Minimum Redundancy - Mutual Information Quotient (MRMR MIQ), were employed to identify the most discriminating features. These features were employed to train 4 machine learning classifiers: Random Forest, Neural Network, Support Vector Machine, and Linear Discriminant Analysis, via 3-fold cross validation.

Results: The four sets of risk categories, and the top Area Under the receiver operating characteristic Curve (AUC) machine classifier performances were: 1) Low ODx and Low mBR grade vs. High ODx and High mBR grade (Low-Low vs. High-High) (AUC = 0.83), 2) Low ODx vs. High ODx (AUC = 0.72), 3) Low ODx vs. Intermediate and High ODx (AUC = 0.58), and 4) Low and Intermediate ODx vs. High ODx (AUC = 0.65). Trained models were tested independent validation set of 53 cases which comprised of Low and High ODx risk, and demonstrated per-patient accuracies ranging from 75 to 86%.

Conclusion: Our results suggest that computerized image analysis of digitized H&E pathology images of early stage ER+ breast cancer might be able predict the corresponding Oncotype DX risk categories.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1186/s12885-018-4448-9DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC5977541PMC
May 2018

High-throughput adaptive sampling for whole-slide histopathology image analysis (HASHI) via convolutional neural networks: Application to invasive breast cancer detection.

PLoS One 2018 24;13(5):e0196828. Epub 2018 May 24.

Dept. of Computing Systems and Industrial Engineering, Universidad Nacional de Colombia, Bogotá, Cundinamarca, Colombia.

Precise detection of invasive cancer on whole-slide images (WSI) is a critical first step in digital pathology tasks of diagnosis and grading. Convolutional neural network (CNN) is the most popular representation learning method for computer vision tasks, which have been successfully applied in digital pathology, including tumor and mitosis detection. However, CNNs are typically only tenable with relatively small image sizes (200 × 200 pixels). Only recently, Fully convolutional networks (FCN) are able to deal with larger image sizes (500 × 500 pixels) for semantic segmentation. Hence, the direct application of CNNs to WSI is not computationally feasible because for a WSI, a CNN would require billions or trillions of parameters. To alleviate this issue, this paper presents a novel method, High-throughput Adaptive Sampling for whole-slide Histopathology Image analysis (HASHI), which involves: i) a new efficient adaptive sampling method based on probability gradient and quasi-Monte Carlo sampling, and, ii) a powerful representation learning classifier based on CNNs. We applied HASHI to automated detection of invasive breast cancer on WSI. HASHI was trained and validated using three different data cohorts involving near 500 cases and then independently tested on 195 studies from The Cancer Genome Atlas. The results show that (1) the adaptive sampling method is an effective strategy to deal with WSI without compromising prediction accuracy by obtaining comparative results of a dense sampling (∼6 million of samples in 24 hours) with far fewer samples (∼2,000 samples in 1 minute), and (2) on an independent test dataset, HASHI is effective and robust to data from multiple sites, scanners, and platforms, achieving an average Dice coefficient of 76%.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0196828PLOS
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC5967747PMC
July 2018

A resolution adaptive deep hierarchical (RADHicaL) learning scheme applied to nuclear segmentation of digital pathology images.

Comput Methods Biomech Biomed Eng Imaging Vis 2018 28;6(3):270-276. Epub 2016 Apr 28.

Department of Biomedical Engineering, Case Western Reserve University, Cleveland, OH, USA.

Deep learning (DL) has recently been successfully applied to a number of image analysis problems. However, DL approaches tend to be inefficient for segmentation on large image data, such as high-resolution digital pathology slide images. For example, typical breast biopsy images scanned at 40× magnification contain billions of pixels, of which usually only a small percentage belong to the class of interest. For a typical naïve deep learning scheme, parsing through and interrogating all the image pixels would represent hundreds if not thousands of hours of compute time using high performance computing environments. In this paper, we present a resolution adaptive deep hierarchical (RADHicaL) learning scheme wherein DL networks at lower resolutions are leveraged to determine if higher levels of magnification, and thus computation, are necessary to provide precise results. We evaluate our approach on a nuclear segmentation task with a cohort of 141 ER+ breast cancer images and show we can reduce computation time on average by about 85%. Expert annotations of 12,000 nuclei across these 141 images were employed for quantitative evaluation of RADHicaL. A head-to-head comparison with a naïve DL approach, operating solely at the highest magnification, yielded the following performance metrics: .9407 vs .9854 Detection Rate, .8218 vs .8489 -score, .8061 vs .8364 true positive rate and .8822 vs 0.8932 positive predictive value. Our performance indices compare favourably with state of the art nuclear segmentation approaches for digital pathology images.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1080/21681163.2016.1141063DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC5935259PMC
April 2016

The lncRNA BORG Drives Breast Cancer Metastasis and Disease Recurrence.

Sci Rep 2017 10 5;7(1):12698. Epub 2017 Oct 5.

Case Comprehensive Cancer Center, Case Western Reserve University, Cleveland, OH, 44106, USA.

Long noncoding RNAs (lncRNAs) have emerged as potent regulators of breast cancer development and progression, including the metastatic spread of disease. Through in silico and biological analyses, we identified a novel lncRNA, BMP/OP-Responsive Gene (BORG), whose expression directly correlates with aggressive breast cancer phenotypes, as well as with metastatic competence and disease recurrence in multiple clinical cohorts. Mechanistically, BORG elicits the metastatic outgrowth of latent breast cancer cells by promoting the localization and transcriptional repressive activity of TRIM28, which binds BORG and induces substantial alterations in carcinoma proliferation and survival. Moreover, inhibiting BORG expression in metastatic breast cancer cells impedes their metastatic colonization of the lungs of mice, implying that BORG acts as a novel driver of the genetic and epigenetic alterations that underlie the acquisition of metastatic and recurrent phenotypes by breast cancer cells.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1038/s41598-017-12716-6DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC5629202PMC
October 2017

Integrated Diagnostics: The Computational Revolution Catalyzing Cross-disciplinary Practices in Radiology, Pathology, and Genomics.

Radiology 2017 10;285(1):12-15

From the Center for Medical Image Science and Visualization, Linköping University Hospital, 581 85 Linköping, Sweden (C.F.L.); Sectra, Linköping, Sweden (C.F.L.); and Departments of Pathology (H.L.G.) and Radiology (P.R.R.), University Hospitals Cleveland Medical Center, Case Western Reserve University, Cleveland, Ohio.

View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1148/radiol.2017170062DOI Listing
October 2017

Intratumoral and peritumoral radiomics for the pretreatment prediction of pathological complete response to neoadjuvant chemotherapy based on breast DCE-MRI.

Breast Cancer Res 2017 05 18;19(1):57. Epub 2017 May 18.

Department of Biomedical Engineering, Case Western Reserve University, Cleveland, OH, 44106, USA.

Background: In this study, we evaluated the ability of radiomic textural analysis of intratumoral and peritumoral regions on pretreatment breast cancer dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) to predict pathological complete response (pCR) to neoadjuvant chemotherapy (NAC).

Methods: A total of 117 patients who had received NAC were retrospectively analyzed. Within the intratumoral and peritumoral regions of T1-weighted contrast-enhanced MRI scans, a total of 99 radiomic textural features were computed at multiple phases. Feature selection was used to identify a set of top pCR-associated features from within a training set (n = 78), which were then used to train multiple machine learning classifiers to predict the likelihood of pCR for a given patient. Classifiers were then independently tested on 39 patients. Experiments were repeated separately among hormone receptor-positive and human epidermal growth factor receptor 2-negative (HR, HER2) and triple-negative or HER2 (TN/HER2) tumors via threefold cross-validation to determine whether receptor status-specific analysis could improve classification performance.

Results: Among all patients, a combined intratumoral and peritumoral radiomic feature set yielded a maximum AUC of 0.78 ± 0.030 within the training set and 0.74 within the independent testing set using a diagonal linear discriminant analysis (DLDA) classifier. Receptor status-specific feature discovery and classification enabled improved prediction of pCR, yielding maximum AUCs of 0.83 ± 0.025 within the HR, HER2 group using DLDA and 0.93 ± 0.018 within the TN/HER2 group using a naive Bayes classifier. In HR, HER2 breast cancers, non-pCR was characterized by elevated peritumoral heterogeneity during initial contrast enhancement. However, TN/HER2 tumors were best characterized by a speckled enhancement pattern within the peritumoral region of nonresponders. Radiomic features were found to strongly predict pCR independent of choice of classifier, suggesting their robustness as response predictors.

Conclusions: Through a combined intratumoral and peritumoral radiomics approach, we could successfully predict pCR to NAC from pretreatment breast DCE-MRI, both with and without a priori knowledge of receptor status. Further, our findings suggest that the radiomic features most predictive of response vary across different receptor subtypes.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1186/s13058-017-0846-1DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC5437672PMC
May 2017

Accurate and reproducible invasive breast cancer detection in whole-slide images: A Deep Learning approach for quantifying tumor extent.

Sci Rep 2017 04 18;7:46450. Epub 2017 Apr 18.

Case Western Reserve University, Cleveland, OH, USA.

With the increasing ability to routinely and rapidly digitize whole slide images with slide scanners, there has been interest in developing computerized image analysis algorithms for automated detection of disease extent from digital pathology images. The manual identification of presence and extent of breast cancer by a pathologist is critical for patient management for tumor staging and assessing treatment response. However, this process is tedious and subject to inter- and intra-reader variability. For computerized methods to be useful as decision support tools, they need to be resilient to data acquired from different sources, different staining and cutting protocols and different scanners. The objective of this study was to evaluate the accuracy and robustness of a deep learning-based method to automatically identify the extent of invasive tumor on digitized images. Here, we present a new method that employs a convolutional neural network for detecting presence of invasive tumor on whole slide images. Our approach involves training the classifier on nearly 400 exemplars from multiple different sites, and scanners, and then independently validating on almost 200 cases from The Cancer Genome Atlas. Our approach yielded a Dice coefficient of 75.86%, a positive predictive value of 71.62% and a negative predictive value of 96.77% in terms of pixel-by-pixel evaluation compared to manually annotated regions of invasive ductal carcinoma.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1038/srep46450DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC5394452PMC
April 2017

A deep learning based strategy for identifying and associating mitotic activity with gene expression derived risk categories in estrogen receptor positive breast cancers.

Cytometry A 2017 06 13;91(6):566-573. Epub 2017 Feb 13.

Biomedical Engineering Department, Case Western Reserve University, Cleveland, Ohio.

The treatment and management of early stage estrogen receptor positive (ER+) breast cancer is hindered by the difficulty in identifying patients who require adjuvant chemotherapy in contrast to those that will respond to hormonal therapy. To distinguish between the more and less aggressive breast tumors, which is a fundamental criterion for the selection of an appropriate treatment plan, Oncotype DX (ODX) and other gene expression tests are typically employed. While informative, these gene expression tests are expensive, tissue destructive, and require specialized facilities. Bloom-Richardson (BR) grade, the common scheme employed in breast cancer grading, has been shown to be correlated with the Oncotype DX risk score. Unfortunately, studies have also shown that the BR grade determined experiences notable inter-observer variability. One of the constituent categories in BR grading is the mitotic index. The goal of this study was to develop a deep learning (DL) classifier to identify mitotic figures from whole slides images of ER+ breast cancer, the hypothesis being that the number of mitoses identified by the DL classifier would correlate with the corresponding Oncotype DX risk categories. The mitosis detector yielded an average F-score of 0.556 in the AMIDA mitosis dataset using a 6-fold validation setup. For a cohort of 174 whole slide images with early stage ER+ breast cancer for which the corresponding Oncotype DX score was available, the distributions of the number of mitoses identified by the DL classifier was found to be significantly different between the high vs low Oncotype DX risk groups (P < 0.01). Comparisons of other risk groups, using both ODX score and histological grade, were also found to present significantly different automated mitoses distributions. Additionally, a support vector machine classifier trained to separate low/high Oncotype DX risk categories using the mitotic count determined by the DL classifier yielded a 83.19% classification accuracy. © 2017 International Society for Advancement of Cytometry.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1002/cyto.a.23065DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC6124660PMC
June 2017

A Deep Convolutional Neural Network for segmenting and classifying epithelial and stromal regions in histopathological images.

Neurocomputing 2016 May 17;191:214-223. Epub 2016 Feb 17.

Department of Biomedical Engineering, Case Western Reserve University, OH 44106, USA.

Epithelial (EP) and stromal (ST) are two types of tissues in histological images. Automated segmentation or classification of EP and ST tissues is important when developing computerized system for analyzing the tumor microenvironment. In this paper, a Deep Convolutional Neural Networks (DCNN) based feature learning is presented to automatically segment or classify EP and ST regions from digitized tumor tissue microarrays (TMAs). Current approaches are based on handcraft feature representation, such as color, texture, and Local Binary Patterns (LBP) in classifying two regions. Compared to handcrafted feature based approaches, which involve task dependent representation, DCNN is an end-to-end feature extractor that may be directly learned from the raw pixel intensity value of EP and ST tissues in a data driven fashion. These high-level features contribute to the construction of a supervised classifier for discriminating the two types of tissues. In this work we compare DCNN based models with three handcraft feature extraction based approaches on two different datasets which consist of 157 Hematoxylin and Eosin (H&E) stained images of breast cancer and 1376 immunohistological (IHC) stained images of colorectal cancer, respectively. The DCNN based feature learning approach was shown to have a F1 classification score of 85%, 89%, and 100%, accuracy (ACC) of 84%, 88%, and 100%, and Matthews Correlation Coefficient (MCC) of 86%, 77%, and 100% on two H&E stained (NKI and VGH) and IHC stained data, respectively. Our DNN based approach was shown to outperform three handcraft feature extraction based approaches in terms of the classification of EP and ST regions.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.neucom.2016.01.034DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC5283391PMC
May 2016