scholarly journals Human-interpretable image features derived from densely mapped cancer pathology slides predict diverse molecular phenotypes

2021 ◽  
Vol 12 (1) ◽  
Author(s):  
James A. Diao ◽  
Jason K. Wang ◽  
Wan Fung Chui ◽  
Victoria Mountain ◽  
Sai Chowdary Gullapally ◽  
...  
Keyword(s):  
Tumor Microenvironment ◽  
Image Features ◽  
Tissue Type ◽  
Type Model ◽  
Biologically Relevant ◽  
Cancer Pathology ◽  
Substantial Progress ◽  
Cancer Types ◽  
Molecular Phenotypes ◽  
Immune Checkpoint Proteins

AbstractComputational methods have made substantial progress in improving the accuracy and throughput of pathology workflows for diagnostic, prognostic, and genomic prediction. Still, lack of interpretability remains a significant barrier to clinical integration. We present an approach for predicting clinically-relevant molecular phenotypes from whole-slide histopathology images using human-interpretable image features (HIFs). Our method leverages >1.6 million annotations from board-certified pathologists across >5700 samples to train deep learning models for cell and tissue classification that can exhaustively map whole-slide images at two and four micron-resolution. Cell- and tissue-type model outputs are combined into 607 HIFs that quantify specific and biologically-relevant characteristics across five cancer types. We demonstrate that these HIFs correlate with well-known markers of the tumor microenvironment and can predict diverse molecular signatures (AUROC 0.601–0.864), including expression of four immune checkpoint proteins and homologous recombination deficiency, with performance comparable to ‘black-box’ methods. Our HIF-based approach provides a comprehensive, quantitative, and interpretable window into the composition and spatial architecture of the tumor microenvironment.

scholarly journals Dense, high-resolution mapping of cells and tissues from pathology images for the interpretable prediction of molecular phenotypes in cancer

2020 ◽  
Author(s):  
James A. Diao ◽  
Wan Fung Chui ◽  
Jason K. Wang ◽  
Richard N. Mitchell ◽  
Sudha K. Rao ◽  
...  
Keyword(s):  
High Resolution ◽  
Image Features ◽  
Tissue Type ◽  
Cell Detection ◽  
Biologically Relevant ◽  
Novel Approach ◽  
Substantial Progress ◽  
Resolution Mapping ◽  
Cancer Types ◽  
Molecular Phenotypes

While computational methods have made substantial progress in improving the accuracy and throughput of pathology workflows for diagnostic, prognostic, and genomic prediction, lack of interpretability remains a significant barrier to clinical integration. In this study, we present a novel approach for predicting clinically-relevant molecular phenotypes from histopathology whole-slide images (WSIs) using human-interpretable image features (HIFs). Our method leverages >1.6 million annotations from board-certified pathologists across >5,700 WSIs to train deep learning models for high-resolution tissue classification and cell detection across entire WSIs in five cancer types. Combining cell- and tissue-type models enables computation of 607 HIFs that comprehensively capture specific and biologically-relevant characteristics of multiple tumors. We demonstrate that these HIFs correlate with well-known markers of the tumor microenvironment (TME) and can predict diverse molecular signatures, including immune checkpoint protein expression and homologous recombination deficiency (HRD). Our HIF-based approach provides a novel, quantitative, and interpretable window into the composition and spatial architecture of the TME.

scholarly journals Recent Advancements on Immunomodulatory Mechanisms of Resveratrol in Tumor Microenvironment

Molecules ◽  
2021 ◽  
Vol 26 (5) ◽  
pp. 1343
Author(s):  
Gagan Chhabra ◽  
Chandra K. Singh ◽  
Deeba Amiri ◽  
Neha Akula ◽  
Nihal Ahmad
Keyword(s):  
Tumor Microenvironment ◽  
Immune Cell ◽  
Cancer Therapies ◽  
Regulatory Cells ◽  
Immunomodulatory Effects ◽  
Cancer Management ◽  
Natural Agents ◽  
Anticancer Effects ◽  
Cancer Types ◽  
Immune Related Genes

Immunomodulation of the tumor microenvironment is emerging as an important area of research for the treatment of cancer patients. Several synthetic and natural agents are being investigated for their ability to enhance the immunogenic responses of immune cells present in the tumor microenvironment to impede tumor cell growth and dissemination. Among them, resveratrol, a stilbenoid found in red grapes and many other natural sources, has been studied extensively. Importantly, resveratrol has been shown to possess activity against various human diseases, including cancer. Mechanistically, resveratrol has been shown to regulate an array of signaling pathways and processes involving oxidative stress, inflammation, apoptosis, and several anticancer effects. Furthermore, recent research suggests that resveratrol can regulate various cellular signaling events including immune cell regulation, cytokines/chemokines secretion, and the expression of several other immune-related genes. In this review, we have summarized recent findings on resveratrol’s effects on immune regulatory cells and associated signaling in various cancer types. Numerous immunomodulatory effects of resveratrol suggest it may be useful in combination with other cancer therapies including immunotherapy for effective cancer management.

scholarly journals The Role of Extracellular Vesicles in the Development of a Cancer Stem Cell Microenvironment Niche and Potential Therapeutic Targets

Cancers ◽  
2021 ◽  
Vol 13 (10) ◽  
pp. 2435
Author(s):  
Thomas J. Brown ◽  
Victoria James
Keyword(s):  
Systematic Review ◽  
Tumor Microenvironment ◽  
Extracellular Vesicles ◽  
Multiple Cancer ◽  
Therapeutic Targeting ◽  
Small Component ◽  
Facilitated Communication ◽  
Cell Microenvironment ◽  
Cancer Types

Cancer stem cells (CSCs) have increasingly been shown to be a crucial element of heterogenous tumors. Although a relatively small component of the population, they increase the resistance to treatment and the likelihood of recurrence. In recent years, it has been shown, across multiple cancer types (e.g., colorectal, breast and prostate), that reciprocal communication between cancer and the microenvironment exists, which is, in part, facilitated by extracellular vesicles (EVs). However, the mechanisms of this method of communication and its influence on CSC populations is less well-understood. Therefore, the aim of this systematic review is to determine the evidence that supports the role of EVs in the manipulation of the tumor microenvironment to promote the survival of CSCs. Embase and PubMed were used to identify all studies on the topic, which were screened using PRISMA guidelines, resulting in the inclusion of 16 studies. These 16 studies reported on the EV content, pathways altered by EVs and therapeutic targeting of CSC through EV-mediated changes to the microenvironment. In conclusion, these studies demonstrated the role of EV-facilitated communication in maintaining CSCs via manipulation of the tumor microenvironment, demonstrating the potential of creating therapeutics to target CSCs. However, further works are needed to fully understand the targetable mechanisms upon which future therapeutics can be based.

scholarly journals Hypoxic tumor microenvironment: Implications for cancer therapy

2020 ◽  
Vol 245 (13) ◽  
pp. 1073-1086
Author(s):  
Sukanya Roy ◽  
Subhashree Kumaravel ◽  
Ankith Sharma ◽  
Camille L Duran ◽  
Kayla J Bayless ◽  
...  
Keyword(s):  
Tumor Microenvironment ◽  
Cancer Cells ◽  
Metabolic Regulation ◽  
Molecular Mechanisms ◽  
Hypoxia Inducible Factor ◽  
Tumor Vasculature ◽  
Therapeutic Strategies ◽  
Therapeutic Resistance ◽  
Low Oxygen ◽  
Cancer Types

Hypoxia or low oxygen concentration in tumor microenvironment has widespread effects ranging from altered angiogenesis and lymphangiogenesis, tumor metabolism, growth, and therapeutic resistance in different cancer types. A large number of these effects are mediated by the transcription factor hypoxia inducible factor 1⍺ (HIF-1⍺) which is activated by hypoxia. HIF1⍺ induces glycolytic genes and reduces mitochondrial respiration rate in hypoxic tumoral regions through modulation of various cells in tumor microenvironment like cancer-associated fibroblasts. Immune evasion driven by HIF-1⍺ further contributes to enhanced survival of cancer cells. By altering drug target expression, metabolic regulation, and oxygen consumption, hypoxia leads to enhanced growth and survival of cancer cells. Tumor cells in hypoxic conditions thus attain aggressive phenotypes and become resistant to chemo- and radio- therapies resulting in higher mortality. While a number of new therapeutic strategies have succeeded in targeting hypoxia, a significant improvement of these needs a more detailed understanding of the various effects and molecular mechanisms regulated by hypoxia and its effects on modulation of the tumor vasculature. This review focuses on the chief hypoxia-driven molecular mechanisms and their impact on therapeutic resistance in tumors that drive an aggressive phenotype. Impact statement Hypoxia contributes to tumor aggressiveness and promotes growth of many solid tumors that are often resistant to conventional therapies. In order to achieve successful therapeutic strategies targeting different cancer types, it is necessary to understand the molecular mechanisms and signaling pathways that are induced by hypoxia. Aberrant tumor vasculature and alterations in cellular metabolism and drug resistance due to hypoxia further confound this problem. This review focuses on the implications of hypoxia in an inflammatory TME and its impact on the signaling and metabolic pathways regulating growth and progression of cancer, along with changes in lymphangiogenic and angiogenic mechanisms. Finally, the overarching role of hypoxia in mediating therapeutic resistance in cancers is discussed.

scholarly journals Synthesis of diagnostic quality cancer pathology images

2020 ◽  
Author(s):  
Adrian B. Levine ◽  
Jason Peng ◽  
David Farnell ◽  
Mitchell Nursey ◽  
Yiping Wang ◽  
...  
Keyword(s):  
Neural Networks ◽  
The Cancer Genome Atlas ◽  
Generative Adversarial Networks ◽  
Diagnostic Systems ◽  
Deep Convolutional Neural Networks ◽  
Diagnostic Quality ◽  
Medical Practitioners ◽  
Cancer Pathology ◽  
Cancer Types ◽  
Synthetic Images

ABSTRACTDeep learning-based computer vision methods have recently made remarkable breakthroughs in the analysis and classification of cancer pathology images. However, there has been relatively little investigation of the utility of deep neural networks to synthesize medical images. In this study, we evaluated the efficacy of generative adversarial networks (GANs) to synthesize high resolution pathology images of ten histological types of cancer, including five cancer types from The Cancer Genome Atlas (TCGA) and the five major histological subtypes of ovarian carcinoma. The quality of these images was assessed using a comprehensive survey of board-certified pathologists (n = 9) and pathology trainees (n = 6). Our results show that the real and synthetic images are classified by histotype with comparable accuracies, and the synthetic images are visually indistinguishable from real images. Furthermore, we trained deep convolutional neural networks (CNNs) to diagnose the different cancer types and determined that the synthetic images perform as well as additional real images when used to supplement a small training set. These findings have important applications in proficiency testing of medical practitioners and quality assurance in clinical laboratories. Furthermore, training of computer-aided diagnostic systems can benefit from synthetic images where labeled datasets are limited (e.g., rare cancers). We have created a publicly available website where clinicians and researchers can attempt questions from the image survey at http://gan.aimlab.ca/.

scholarly journals TISCH: a comprehensive web resource enabling interactive single-cell transcriptome visualization of tumor microenvironment

2020 ◽  
Vol 49 (D1) ◽  
pp. D1420-D1430
Author(s):  
Dongqing Sun ◽  
Jin Wang ◽  
Ya Han ◽  
Xin Dong ◽  
Jun Ge ◽  
...  
Keyword(s):  
Gene Expression ◽  
Tumor Microenvironment ◽  
Single Cell ◽  
Large Scale ◽  
Differential Expression Analysis ◽  
Enrichment Analysis ◽  
Functional Enrichment ◽  
Multiple Cancer ◽  
Web Resource ◽  
Cancer Types

Abstract Cancer immunotherapy targeting co-inhibitory pathways by checkpoint blockade shows remarkable efficacy in a variety of cancer types. However, only a minority of patients respond to treatment due to the stochastic heterogeneity of tumor microenvironment (TME). Recent advances in single-cell RNA-seq technologies enabled comprehensive characterization of the immune system heterogeneity in tumors but posed computational challenges on integrating and utilizing the massive published datasets to inform immunotherapy. Here, we present Tumor Immune Single Cell Hub (TISCH, http://tisch.comp-genomics.org), a large-scale curated database that integrates single-cell transcriptomic profiles of nearly 2 million cells from 76 high-quality tumor datasets across 27 cancer types. All the data were uniformly processed with a standardized workflow, including quality control, batch effect removal, clustering, cell-type annotation, malignant cell classification, differential expression analysis and functional enrichment analysis. TISCH provides interactive gene expression visualization across multiple datasets at the single-cell level or cluster level, allowing systematic comparison between different cell-types, patients, tissue origins, treatment and response groups, and even different cancer-types. In summary, TISCH provides a user-friendly interface for systematically visualizing, searching and downloading gene expression atlas in the TME from multiple cancer types, enabling fast, flexible and comprehensive exploration of the TME.

scholarly journals Prognostic Indications of Elevated MCT4 and CD147 across Cancer Types: A Meta-Analysis

2015 ◽  
Vol 2015 ◽  
pp. 1-14 ◽  
Author(s):  
Cory D. Bovenzi ◽  
James Hamilton ◽  
Patrick Tassone ◽  
Jennifer Johnson ◽  
David M. Cognetti ◽  
...  
Keyword(s):  
Overall Survival ◽  
Tumor Microenvironment ◽  
Cancer Cells ◽  
Meta Analysis ◽  
Critical Role ◽  
Disease Free Survival ◽  
Free Survival ◽  
Cd147 Expression ◽  
Cancer Types ◽  
Disease Free

Background. Metabolism in the tumor microenvironment can play a critical role in tumorigenesis and tumor aggression. Metabolic coupling may occur between tumor compartments; this phenomenon can be prognostically significant and may be conserved across tumor types. Monocarboxylate transporters (MCTs) play an integral role in cellular metabolism via lactate transport and have been implicated in metabolic synergy in tumors. The transporters MCT1 and MCT4 are regulated via expression of their chaperone, CD147.Methods. We conducted a meta-analysis of existing publications on the relationship between MCT1, MCT4, and CD147 expression and overall survival and disease-free survival in cancer, using hazard ratios derived via multivariate Cox regression analyses.Results. Increased MCT4 expressions in the tumor microenvironment, cancer cells, or stromal cells were all associated with decreased overall survival and decreased disease-free survival (p<0.001for all analyses). Increased CD147 expression in cancer cells was associated with decreased overall survival and disease-free survival (p<0.0001for both analyses). Few studies were available on MCT1 expression; MCT1 expression was not clearly associated with overall or disease-free survival.Conclusion. MCT4 and CD147 expression correlate with worse prognosis across many cancer types. These results warrant further investigation of these associations.

scholarly journals Mechanism- and Immune Landscape-Based Ranking of Therapeutic Responsiveness of 22 Major Human Cancers to Next Generation Anti-CTLA-4 Antibodies

Cancers ◽  
2020 ◽  
Vol 12 (2) ◽  
pp. 284
Author(s):  
Peng Zhang ◽  
Xinxin Xiong ◽  
Christian Rolfo ◽  
Xuexiang Du ◽  
Yan Zhang ◽  
...  
Keyword(s):  
T Cells ◽  
Tumor Microenvironment ◽  
Lung Carcinoma ◽  
Small Cell Lung Carcinoma ◽  
Clinical Development ◽  
Small Cell ◽  
Small Cell Lung ◽  
Cell Lung Carcinoma ◽  
Cancer Types ◽  
New Generation

Background: CTLA-4 was the first immune checkpoint targeted for cancer therapy and the first target validated by the FDA (Food and Drug Administration) after approval of the anti-CTLA-4 antibody, Ipilimumab. However, clinical response rates to anti-CTLA-4 antibodies are lower while the rates of immunotherapy-related adverse events (irAE) are higher than with anti-PD-1 antibodies. As a result, the effort to target CTLA-4 for cancer immunotherapy has stagnated. To reinvigorate CTLA-4-targeted immunotherapy, we and others have reported that rather than blocking CTLA-4 interaction with its cognate targets, CD80 and CD86, anti-CTLA-4 antibodies achieve their therapeutic responses through selective depletion of regulatory T cells in the tumor microenvironment. Accordingly, we have developed a new generation of anti-CTLA-4 antibodies with reduced irAE and enhanced antibody-dependent cell-mediated cytotoxicity/phagocytosis (ADCC/ADCP). A major unresolved issue is how to select appropriate cancer types for future clinical development. Methods: We generated a landscape of the immune tumor microenvironment from RNAseq and genomic data of 7279 independent cancer samples belonging to 22 cancer types from The Cancer Genomics Atlas (TCGA) database. Based primarily on genomic and RNAseq data from pre-treatment clinical samples of melanoma patients who were later identified as responders and nonresponders to the anti-CTLA-4 antibody Ipilimumab, we identified 5 ranking components of responsiveness to anti-CTLA-4, including CTLA-4 gene expression, ADCC potential, mutation burden, as well as gene enrichment and cellular composition that favor CTLA-4 responsiveness. The total ranking number was calculated by the sum of 5 independent partitioning values, each comprised of 1–3 components. Results: Our analyses predict metastatic melanoma as the most responsive cancer, as expected. Surprisingly, non-small cell lung carcinoma (NSCLC) is predicted to be highly responsive to anti-CTLA-4 antibodies. Single-cell RNAseq analysis and flow cytometry of human NSCLC-infiltrating T cells supports the potential of anti-CTLA-4 antibodies to selectively deplete intratumoral Treg. Conclusions: Our in silico and experimental analyses suggest that non-small cell lung carcinoma will likely respond to a new generation of anti-CTLA-4 monoclonal antibodies. Our approach provides an objective ranking of the sensitivity of human cancers to anti-CTLA-4 antibodies. The comprehensive ranking of major cancer types provides a roadmap for clinical development of the next generation of anti-CTLA-4 antibodies.

Sign in / Sign up

Export Citation Format

Share Document