PanCancer insights from The Cancer Genome Atlas: the pathologist's perspective

Abstract

The Cancer Genome Atlas (TCGA) represents one of several international consortia dedicated to performing comprehensive genomic and epigenomic analyses of selected tumour types to advance our understanding of disease and provide an open-access resource for worldwide cancer research. Thirty-three tumour types (selected by histology or tissue of origin, to include both common and rare diseases), comprising >11 000 specimens, were subjected to DNA sequencing, copy number and methylation analysis, and transcriptomic, proteomic and histological evaluation. Each cancer type was analysed individually to identify tissue-specific alterations, and make correlations across different molecular platforms. The final dataset was then normalized and combined for the PanCancer Initiative, which seeks to identify commonalities across different cancer types or cells of origin/lineage, or within anatomically or morphologically related groups. An important resource generated along with the rich molecular studies is an extensive digital pathology slide archive, composed of frozen section tissue directly related to the tissues analysed as part of TCGA, and representative formalin-fixed paraffin-embedded, haematoxylin and eosin (H&E)-stained diagnostic slides. These H&E image resources have primarily been used to verify diagnoses and histological subtypes with some limited extraction of standard pathological variables such as mitotic activity, grade, and lymphocytic infiltrates. Largely overlooked is the richness of these scanned images for more sophisticated feature extraction approaches coupled with machine learning, and ultimately correlation with molecular features and clinical endpoints. Here, we document initial attempts to exploit TCGA imaging archives, and describe some of the tools, and the rapidly evolving image analysis/feature extraction landscape. Our hope is to inform, and ultimately inspire and challenge, the pathology and cancer research communities to exploit these imaging resources so that the full potential of this integral platform of TCGA can be used to complement and enhance the insightful integrated analyses from the genomic and epigenomic platforms.

Accurate Diagnosis of Colorectal Cancer Based on Histopathology Images Using Artificial Intelligence

“…The hospitals involved are located in the major metropolitan areas of China serving >139 million population, including those most prestigious hospitals in pathology in China: XH, FUS, CGH, SWH, and AMU; other state-level esteemed hospitals: SYU, NJD, GPH, HPH, and TXH; and a regional reputable PCH. All WSIs were from FFPE tissues, except part of TCGA WSIs were from frozen tissues 27 . The process of collection, quality control, and digitalization of the WSIs was described in Supplementary-Text 1.a.…”

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

“…From these base-level results, integrative analyses were performed to derive six immune subtypes, spanning tumor types and subtypes. The comprehensive pairing of clinical, sample, molecular tumor, and immune characterizations with H&E WSIs in the TCGA is a unique resource (Cooper et al, 2017) and offers the possibility of identifying relationships between computational staining of whole-slide images and other measures of immune response that may in turn inform research into immuno-oncological therapy. In this work, we characterize spatial patterns of TILs and present relationships between TIL patterns and immune subtypes, tumor types, immune cell fractions, and patient survival, illustrating the potential of this kind of analysis and the kinds of questions that can be explored.…”

Deep learning-based cross-classifications reveal conserved spatial behaviors within tumor histological images

“…Image datasets for hematoxylin and eosin (H&E), immunohistochemistry (IHC), and spatial -omic imaging technologies are rapidly growing (Litjens et al 2017). Improved computational approaches for analyzing cancer images would therefore be valuable, not only for traditional tasks such as histopathological classification and cell segmentation (Cooper et al 2018), but also for novel questions such as the de novo discovery of spatial patterns that distinguish cancer types. The search for recurrent spatial patterns is analogous to the search for common driver mutations or expression signatures based on cancer sequencing (Bailey et al 2018), yet this paradigm has been little explored for cancer image data.…”