Search

Tissue-based cancer studies can generate large amounts of histology data in the form of glass slides. These slides contain important diagnostic, prognostic, and biological information and can be digitized into expansive and high-resolution whole-slide images using slide-scanning devices. Effectively utilizing digital pathology data in cancer research requires the ability to manage, visualize, share, and perform quantitative analysis on these large amounts of image data, tasks that are often complex and difficult for investigators with the current state of commercial digital pathology software. In this article, we describe the Digital Slide Archive (DSA), an open-source web-based platform for digital pathology. DSA allows investigators to manage large collections of histologic images and integrate them with clinical and genomic metadata. The open-source model enables DSA to be extended to provide additional capabilities. Cancer Res; 77(21); e75–78. ©2017 AACR.

Whole-slide histology images contain information that is valuable for clinical and basic science investigations of cancer but extracting quantitative measurements from these images is challenging for researchers who are not image analysis specialists. In this article, we describe HistomicsML2, a software tool for learn-by-example training of machine learning classifiers for histologic patterns in whole-slide images. This tool improves training efficiency and classifier performance by guiding users to the most informative training examples for labeling and can be used to develop classifiers for prospective application or as a rapid annotation tool that is adaptable to different cancer types. HistomicsML2 runs as a containerized server application that provides web-based user interfaces for classifier training, validation, exporting inference results, and collaborative review, and that can be deployed on GPU servers or cloud platforms. We demonstrate the utility of this tool by using it to classify tumor-infiltrating lymphocytes in breast carcinoma and cutaneous melanoma.Significance:An interactive machine learning tool for analyzing digital pathology images enables cancer researchers to apply this tool to measure histologic patterns for clinical and basic science studies.

Supplementary tables.

Supplementary figures and table legends.

Software preview video.

Video

Microscopic examination of pathology slides is essential to disease diagnosis and biomedical research; however, traditional manual examination of tissue slides is laborious and subjective. Tumor whole-slide image (WSI) scanning is becoming part of routine clinical procedure and produces massive data that capture tumor histological details at high resolution. Furthermore, the rapid development of deep learning algorithms has significantly increased the efficiency and accuracy of pathology image analysis. In light of this progress, digital pathology is fast becoming a powerful tool to assist pathologists.Studying tumor tissue and its surrounding microenvironment provides critical insight into tumor initiation, progression, metastasis, and potential therapeutic targets. Nuclei segmentation and classification are critical to pathology image analysis, especially in characterizing and quantifying the tumor microenvironment (TME). Computational algorithms have been developed for nuclei segmentation and TME quantification within image patches; however, existing algorithms are computationally intensive and time-consuming for WSI analysis.In this study, we present Histology-based Detection using Yolo (HD-Yolo), a new method that significantly accelerates nuclei segmentation and TME quantification. We demonstrate that HD-Yolo outperforms existing methods for WSI analysis in nuclei detection and classification accuracy, as well as computation time.

The factors driving therapy resistance in diffuse glioma remain poorly understood. To identify treatment-associated cellular and genetic changes, we analyzed RNA and/or DNA sequencing data from the temporally separated tumor pairs of 304 adult patients with isocitrate dehydrogenase (IDH)-wild-type and IDH-mutant glioma. Tumors recurred in distinct manners that were dependent on IDH mutation status and attributable to changes in histological feature composition, somatic alterations, and microenvironment interactions. Hypermutation and acquired CDKN2A deletions were associated with an increase in proliferating neoplastic cells at recurrence in both glioma subtypes, reflecting active tumor growth. IDH-wild-type tumors were more invasive at recurrence, and their neoplastic cells exhibited increased expression of neuronal signaling programs that reflected a possible role for neuronal interactions in promoting glioma progression. Mesenchymal transition was associated with the presence of a myeloid cell state defined by specific ligand-receptor interactions with neoplastic cells. Collectively, these recurrence-associated phenotypes represent potential targets to alter disease progression.

Pathologic diagnosis of bone marrow disorders relies in part on microscopic analysis of bone marrow aspirate (BMA) smears and manual counting of marrow nucleated cells to obtain a differential cell count (DCC). This manual process has significant limitations, including analysis of only a small subset of optimal slide areas and nucleated cells, and inter-observer variability due to differences in cell selection and classification. To address these shortcomings, we developed an automated machine learning-based pipeline for obtaining 11-component DCCs on whole-slide BMAs. This pipeline utilizes a sequential process of identifying optimal BMA regions with high proportions of marrow nucleated cells, detecting individual cells within these optimal areas, and classifying these cells into one of 11 DCC components. Convolutional neural network models were trained on 396,048 BMA region, 28,914 cell boundary, and 1,510,976 cell class images from manual annotations. The resulting automated pipeline produces 11-component DCCs that demonstrate high statistical and diagnostic concordance with manual DCCs among a heterogeneous group of testing BMA slides with varying pathologies and cellularities. Additionally, we show that automated analysis can reduce intra-slide variance in DCCs by analyzing the whole slide and marrow nucleated cells within optimal regions. Finally, pipeline outputs of region classification, cell detection, and cell classification can be visualized using whole-slide image analysis software. This study demonstrates the feasibility of a fully-automated pipeline for generating DCCs on scanned whole-slide BMA images, with the potential for improving the current standard of practice for utilizing BMA smears in the laboratory analysis of hematologic disorders.

Tumor-Infiltrating Lymphocytes (TILs) have strong prognostic and predictive value in breast cancer, but their visual assessment is subjective. To improve reproducibility, the International Immuno-oncology Working Group recently released recommendations for the computational assessment of TILs that build on visual scoring guidelines. However, existing resources do not adequately address these recommendations due to the lack of annotation datasets that enable joint, panoptic segmentation of tissue regions and cells. Moreover, existing deep-learning architectures focus entirely on either tissue segmentation or object detection, which complicates the process of TILs assessment by necessitating the use of multiple models with inconsistent predictions. We introducePanopTILs, a region and cell-level annotation dataset containing 814,886 nuclei from 151 patients, openly accessible at:sites.google.com/view/panoptils. PanopTILs enabled us to developMuTILs, a convolutional neural network architecture optimized for assessing TILs in accordance with clinical recommendations. MuTILs is a concept bottleneck model designed to be interpretable and to encourage sensible predictions at multiple resolutions. Using a rigorous internal-external cross-validation procedure, MuTILs achieves an AUROC of 0.93 for lymphocyte detection and a DICE coefficient of 0.81 for tumor-associated stroma segmentation. Our computational score closely matched visual scores (Spearman R=0.58, p

Comprehensive sequencing of patient tumors reveals genomic mutations across tumor types that enable tumorigenesis and progression. A subset of oncogenic driver mutations results in neomorphic activity where the mutant protein mediates functions not engaged by the parental molecule. Here, we identify prevalent variant-enabled neomorph-protein-protein interactions (neoPPI) with a quantitative High Throughput differential Screening (qHT-dS) platform. Coupling of highly sensitive BRET biosensors with miniaturized co-expression in an ultra-HTS format allows large-scale monitoring of interactions of wild-type and mutant variant counterparts with a library of cancer-associated proteins in live cells. Screening of 17,792 interactions with 2,172,864 data points revealed a landscape of gain-of-interactions encompassing both oncogenic and tumor suppressor mutations. For example, the recurrent BRAF V600E lesion mediates KEAP1 neoPPI, rewiring a BRAF(V600E)/KEAP1 signaling axis and creating collateral vulnerability to NQO1 substrates, offering a combination therapeutic strategy. Thus, cancer genomic alterations can create neo-interactions, informing variant-directed therapeutic approaches for precision medicine.