Utilizing Automated Breast Cancer Detection to Identify Spatial Distributions of Tumor-Infiltrating Lymphocytes in Invasive Breast Cancer.

Quantitative assessment of spatial relations between tumor and tumor-infiltrating lymphocytes (TIL) is increasingly important in both basic science and clinical aspects of breast cancer research. We have developed and evaluated convolutional neural network analysis pipelines to generate combined maps of cancer regions and TILs in routine diagnostic breast cancer whole slide tissue images. The combined maps provide insight about the structural patterns and spatial distribution of lymphocytic infiltrates and facilitate improved quantification of TILs. Both tumor and TIL analyses were evaluated by using three convolutional neural network networks (34-layer ResNet, 16-layer VGG, and Inception v4); the results compared favorably with those obtained by using the best published methods. We have produced open-source tools and a public data set consisting of tumor/TIL maps for 1090 invasive breast cancer images from The Cancer Genome Atlas. The maps can be downloaded for further downstream analyses.

Identification of breast cancer margins using intraoperative near-infrared imaging.

BACKGROUND AND OBJECTIVES: Current methods of intraoperative breast cancer margin assessment are labor intensive, not fully reliable, and time consuming; therefore novel strategies are necessary. We hypothesized that near infrared (NIR) intraoperative molecular imaging using systemic indocyanine green (ICG) would be helpful in discerning tumor margins. METHODS: A mammary cancer cell line, 4T1, was used to establish tumors in mouse flanks (n = 60). Tumors were excised 24 hr after intravenous ICG. Assessment of residual tumor in the wound bed was performed using a combination of NIR imaging and traditional method (by visual inspection and palpation) versus traditional method alone. Next we performed a clinical trial to evaluate the role of NIR imaging after systemic ICG for the margin assessment of 12 patients undergoing breast-conserving surgery. RESULTS: Traditional margin assessment identified 30% of positive margins while NIR imaging identified 90% of positive margins. In our clinical trial, all tumors were detected by NIR imaging and there was fluorescent evidence of residual tumor in the tumor bed in 6 of the 12 patients. None of these patients had positive margins on pathology. CONCLUSIONS: Systemic ICG reliably accumulates in breast cancers in murine models as well as human breast cancer. While NIR imaging is helpful for detection of retained tumor margins in our animal model, intraoperative imaging for precise margin detection will need further refinement before clinical value can be obtained. J. Surg. Oncol. 2016;113:508-514. © 2016 Wiley Periodicals, Inc.

Juvenile Granulosa Cell Tumor Mimicking HAIR-AN in a 4-Year-Old: A Case Report.

Predominantly androgen secreting juvenile granulosa cell tumors (JGCT) are uncommon and few reports exist in the literature. We present a case of a JGCT which presented with signs of prepubertal hyperandrogenism and insulin resistance to highlight the possible interaction between hyperandrogenemia and hyperinsulinism. We conducted chart review of a rare androgen secreting JGCT accompanied by hyperinsulinemia in a prepubertal patient. A 4-year-old girl presented with acanthosis nigricans and hyperinsulinism mimicking the Hyperandrogenism Insulin Resistance and Acanthosis Nigricans (HAIR-AN) syndrome at an age much younger than is typical for this diagnosis. Laboratory studies revealed elevated insulin, inhibin A and B, and total testosterone. All laboratory results normalized after unilateral salpingo-oophorectomy. The final diagnosis was Stage IA JGCT. This case highlights the importance of including ovarian tumors in the differential diagnosis when considering causes of virilization and insulin resistance. Our case illustrates the potential relationship between excess testosterone secretion and hyperinsulinemia and strengthens evidence that hyperandrogenemia may promote hyperinsulinism in ovarian disease.

Clinical pathology resident education during the COVID-19 pandemic.

COVID-19 arrived at our medical centre in March 2020 with substantial force. Clinical pathology concepts began to have a new, direct relevance to our residents' lives. As we wondered 'Have I been exposed? Do I need to self-isolate? Are the tests reliable? Am I protecting myself adequately while handling specimens?', these questions drew new interest in laboratory methods, test interpretation and limitations, supply chain issues, safety and quality. By incorporating SARS-CoV-2 teaching points into laboratory medicine lectures, we enlivened concepts of sensitivity, specificity, predictive value and methodologic issues in serologic, molecular and antigen testing for pathology residents. We drew from the emerging literature on SARS-CoV-2 to create lectures and added details from our own institutional experience with COVID-19. When the pandemic fades from memory, clinical pathology education can still benefit from mnemonics, analogies, anecdotes and creative efforts that capture the attention of the audience.

Deep Learning-Based Mapping of Tumor Infiltrating Lymphocytes in Whole Slide Images of 23 Types of Cancer.

The role of tumor infiltrating lymphocytes (TILs) as a biomarker to predict disease progression and clinical outcomes has generated tremendous interest in translational cancer research. We present an updated and enhanced deep learning workflow to classify 50x50 um tiled image patches (100x100 pixels at 20x magnification) as TIL positive or negative based on the presence of 2 or more TILs in gigapixel whole slide images (WSIs) from the Cancer Genome Atlas (TCGA). This workflow generates TIL maps to study the abundance and spatial distribution of TILs in 23 different types of cancer. We trained three state-of-the-art, popular convolutional neural network (CNN) architectures (namely VGG16, Inception-V4, and ResNet-34) with a large volume of training data, which combined manual annotations from pathologists (strong annotations) and computer-generated labels from our previously reported first-generation TIL model for 13 cancer types (model-generated annotations). Specifically, this training dataset contains TIL positive and negative patches from cancers in additional organ sites and curated data to help improve algorithmic performance by decreasing known false positives and false negatives. Our new TIL workflow also incorporates automated thresholding to convert model predictions into binary classifications to generate TIL maps. The new TIL models all achieve better performance with improvements of up to 13% in accuracy and 15% in F-score. We report these new TIL models and a curated dataset of TIL maps, referred to as TIL-Maps-23, for 7983 WSIs spanning 23 types of cancer with complex and diverse visual appearances, which will be publicly available along with the code to evaluate performance. Code Available at: https://github.com/ShahiraAbousamra/til_classification.

Comparison of Different Classifiers with Active Learning to Support Quality Control in Nucleus Segmentation in Pathology Images.

Segmentation of nuclei in whole slide tissue images is a common methodology in pathology image analysis. Most segmentation algorithms are sensitive to input algorithm parameters and the characteristics of input images (tissue morphology, staining, etc.). Because there can be large variability in the color, texture, and morphology of tissues within and across cancer types (heterogeneity can exist even within a tissue specimen), it is likely that a set of input parameters will not perform well across multiple images. It is, therefore, highly desired, and necessary in some cases, to carry out a quality control of segmentation results. This work investigates the application of machine learning in this process. We report on the application of active learning for segmentation quality assessment for pathology images and compare three classification methods, Support Vector Machine (SVM), Random Forest (RF) and Convolutional Neural Network (CNN), for their performance improvement and efficiency.

Spatial Organization and Molecular Correlation of Tumor-Infiltrating Lymphocytes Using Deep Learning on Pathology Images.

Beyond sample curation and basic pathologic characterization, the digitized H&E-stained images of TCGA samples remain underutilized. To highlight this resource, we present mappings of tumor-infiltrating lymphocytes (TILs) based on H&E images from 13 TCGA tumor types. These TIL maps are derived through computational staining using a convolutional neural network trained to classify patches of images. Affinity propagation revealed local spatial structure in TIL patterns and correlation with overall survival. TIL map structural patterns were grouped using standard histopathological parameters. These patterns are enriched in particular T cell subpopulations derived from molecular measures. TIL densities and spatial structure were differentially enriched among tumor types, immune subtypes, and tumor molecular subtypes, implying that spatial infiltrate state could reflect particular tumor cell aberration states. Obtaining spatial lymphocytic patterns linked to the rich genomic characterization of TCGA samples demonstrates one use for the TCGA image archives with insights into the tumor-immune microenvironment.