Introduction: Liver biopsy is the gold standard to diagnose non-alcoholic Fatty Liver Disease (NAFLD) and the more progressive form steatohepatitis (NASH), and is the primary outcome in clinical trials for NAFLD treatment. The diagnosis is confirmed using the histopathological NAFLD-activity score (NAS), which grades the severity of steatosis, hepatocellular ballooning degeneration, and inflammation. Histopathological disease scoring systems are, however, subjective and prone to inter- and intra-observer variation. We therefore apply a deep learning image analysis strategy to obtain a more accurate and objective method for staging NAS in mouse models of NAFLD.

Materials and Methods: Using the Artificial Intelligence (AI) Deep Learning software from Visiopharm, the strategy was to perform a segmentation of liver biopsy sections stained for H&E from diet-induced obese (DIO) NASH mouse models and chow-fed controls. Inflammatory cells, lipid droplets, ballooned hepatocytes and hepatocytes with and without steatosis were annotated in H&E sections. Moreover, central veins and portal tracts were annotated. Thereafter, segmentations were postprocessed into the corresponding histopathological scores, and in addition quantitative measures of inflammation and steatosis were obtained.

Results: AI Deep Learning applications successfully recognized inflammatory cells, hepatocytes with and without steatosis, and ballooned hepatocytes in DIO-NASH mice. The app performance was comparable to manual evaluation of NAS, and confirmed significant improvement of NAS after 12w treatment with semaglutide in a DIO-NASH mouse model.

Discussion and Conclusion: We here demonstrate a deep-learning based approach to obtain the NAS scores in translational obese mouse models of DIO-NASH. A deep-learning approach for pattern recognition allows rapid and reproducible quantification of histological NASH parameters.

Agnete Overgaard, Research Scientist, Gubra

Agnete Overgaard has worked with Gubra for the past three years as a research scientist in the Tissue Research department. In this role she is responsible for processes ranging from optimizing protein and mRNA visualisation in tissue preparations, all the way to image analysis and histopathological evaluation of biopsy preparations from Gubra’s diverse animal models. Gubra is a biotech company that offers contract research services within the metabolic space, including extensive histology data packages. Over the years she has been involved in digitalization of workflows from biopsy to graph, including the development of several new image analyses. Recently, she was involved in the development of deep learning-based histopathological scoring of non-alcoholic fatty liver disease, that enables objective evaluation along with quantitative outputs.

She was previously a postdoc within the neurobiology field at University of British Columbia and Copenhagen University Hospital Rigshospitalet, and she holds a PhD in Neuroscience from the University of Copenhagen.

The immune response is spatially and temporally regulated. The density and location of immune cells in the tumor microenvironment (TME) have important diagnostic and prognostic values. Single cell-based multiomic technologies have exponentially increased our understanding of the numerous cellular and molecular networks regulating tumor initiation and progression. However, these techniques do not provide information about the spatial organization of cells or cell-cell interactions. Affordable, accessible, and easy to execute multiplexing techniques that allow spatial resolution of immune cells in tissue sections are needed to complement single cell-based high-throughput technologies.

We have developed a strategy that integrates serial imaging, sequential labeling, and image alignment to generate virtual multiparameter slides of whole tissue sections. Virtual slides are subsequently analyzed in an automated fashion using the VIS software allowing us to identify, quantify, and map cell populations of interest. Specifically, the image analysis is performed using the analysis modules Tissuealign, Author, and HISTOmap. Here, we propose a strategy for the rational design of tissue multiplex assays using commercially available reagents, affordable microscopy equipment, and user-friendly software. Using this strategy, we created one virtual slide comprising 11 biomarkers plus two frequently used histological stains: hematoxylin and eosin (H&E) and picrosirius red (PSR). Multiple immune cell populations were identified, located, and quantified in different tissue compartments and their spatial distribution resolved using tissue heatmaps. This strategy maximizes the information that can be gained from limited clinical specimens and is applicable to formalin-fixed paraffin-embedded (FFPE) archived tissue samples, including whole tissue, core needle biopsies, and tissue microarrays. We propose this methodology as a useful guide for designing custom assays for identification, quantification, and mapping of immune cell populations in the TME.

Presented on March 22, 2021 at an XTalks webinar.

• • Integration of serial imaging, sequential labeling, and image alignment in the experimental design of imaging assays can greatly increase the number of markers that can be visualized simultaneously, expand the possibilities of the analysis, and extract more information from precious clinical specimens.

• • Virtual multiplexing allows to determine how markers visualized in one section spatially relate to markers visualized in another contiguous section.

• • The use of whole tissue sections instead of selected fields of view for the analysis, results in an unbiased representation of the TME.

• • The use of tissue heatmaps greatly simplify the visual representation of the spatial organization of cells in the tissue.

Manuel Flores, Ph.D. Candidate

Manuel Flores obtained his Biochemistry degree from Havana University, Cuba. In 2015 Manuel enrolled in the Immunology and Virology Master Program at Université de Montréal (Canada) and fast tracked to the Immunology and Virology PhD Program in 2016.

His doctoral research project focuses on characterizing the liver resident and infiltrating immune cell populations and their role in the pathogenesis of chronic liver diseases due to persistent viral and toxic injuries, including fibrosis and hepatocellular carcinoma. His research interests center around the spatial organization of immune cells in the hepatic tissue microenvironment, and the delineation of the multiple cell-cell interactions and their respective biological significance in health and disease. Manuel Flores is the recipient of doctoral scholarships from University of Montreal and from Fonds de recherche du Québec – Santé (FRQS).

90% of breast cancer mortality is caused by distant metastasis, a process that involves both dissemination of cancer cells to distant sites as well as their proliferation after arrival. However, prognostic assays currently used in the clinic are based on proliferation and do not measure tumor cell dissemination potential. Based on intravital imaging of tumor cell dissemination in live animals, we previously reported two biomarkers for the development of distant metastasis: TMEM Score and MenaCalc. TMEM Score is based upon the density of Tumor Microenvironment of Metastasis (TMEM) doorways, portals for cancer cell intravasation and dissemination formed by the confluence of a Mena overexpressing tumor cell, a pro-angiogenic macrophage, and an endothelial cell. MenaCalc is a pattern of expression of the actin-regulatory protein Mena which leads tumor cells to undergo epithelial-to-mesenchymal transition (EMT) and become highly motile.

Using digital pathology, we analytically validated an automated analysis of TMEM doorways that reduced pathologist time by an order of magnitude and enabled the rapid clinical validation of TMEM Score. While TMEM score has been validated for prognosticating metastatic outcome in HR+/HER2- patients, statistical significance was not observed in patients with triple negative or HER2+ breast cancers. Furthermore, MenaCalc has been shown to be prognostic in some cohorts of patients with triple negative disease but the prognostic value of MenaCalc in HR+ disease is still unclear. Since TMEM doorways and MenaCalc are mechanistically linked (but independent) biomarkers, we investigated if a combined TMEM-MenaCalc biomarker can improve the prognostication ability of either biomarker alone. Again, using digital pathology, we evaluated several different methods of combining TMEM and MenaCalc scores to create a multiparameter quantitative analysis with dramatically improved prognostic ability for distant metastasis in breast cancer patients.

Presented as a LabRoots webinar on February 9, 2021.

• • Discuss the process of tumor cell dissemination from breast tumors.

• • Summarize how live imaging studies are able to discover mechanisms of metastasis which can then be turned into prognostic biomarkers for metastasis.

• • Outline how digital whole slide scanning and digital pathology can be used to automate as well as analytically and clinically validate prognostic biomarkers.

David Entenberg, PhD, Assistant Professor, Department of Anatomy and Structural Biology, Albert Einstein College of Medicine

Dr. David Entenberg received his bachelor’s degree in physics at SUNY Stony Brook and his Masters’ degree in chemical physics from Weizmann Institute of Science in Israel. He then obtained his Ph.D. in cell biology from the University of Kent. In 2006 he brought his imaging expertise to Albert Einstein College of Medicine and now serves as the Director of Technology Development for Einstein’s Gruss Lipper Biophotonics Center and its associated Integrated Imaging Program. He also leads the development of novel and innovative imaging techniques that provide high-resolution visualization of the tumor microenvironment.

Blood vessels bring oxygen and nutrients to every cell in the body while removing waste and allowing immune cells to survey. They do the same in cancer and other diseases. In most types of tumors, new vessels produced through angiogenesis have abnormal structure and function, leading to impaired perfusion that paradoxically supports malignancy. For this reason, the study of the micro-anatomy, morphology and function of blood vessels in tumors is essential to find new vulnerabilities to be targeted in our fight to tumors.

Dr Seano will present an overview of his works on blood vessel histological investigation. The study of the abnormalities of tumor blood vessels elucidated tumor features and is still shedding light on its pathology. Tumor vessels may be characterized by abnormal morphology, disrupted vascular basement membrane and reduced pericyte coverage. This causes dysfunction in perfusion and consequently leads to an hypoxic and immunosuppressive environment, typical of tumors. Breast cancer and glioblastoma are two of the most important tumors in the field of vascular microenvironment since we learnt that we can therapeutically modulate and temporally normalize vascular function. Dr Seano will show digital pathology results published during his period at Harvard Medical School and unpublished data from his own new lab.

Presented at the virtual Pathology Visions 2020 meeting during Visiopharm’s Industry Workshop on Monday, October 26, 2020.

Dr Giorgio Seano, Head of the Tumor Microenvironment LabInstitut Curie, Orsay-Paris (France)

Dr Seano is the Head of the Tumor Microenvironment Lab at Institut Curie, Orsay-Paris (France). His scientific interests are tumor vasculature, vessel co-option, cell migration and radioresistance. Among others, he published on Science, Cancer Cell, PNAS, Nat Cell Biol, Nat Biom Eng, JNCI, Blood and Cancer Res.

Are you getting the maximum out of your sample images?

In this webinar, Joseph Daniele Ph.D. will demonstrate the power of deep learning to investigate and quantify variability within a cell.

Determining mitotic activity is a common component of many tumor grading systems but relies on tedious identification and enumeration of mitotic figures within selected fields of view. Using Visiopharm’s deep learning module, Joe and his group have trained a classifier to automatically identify and count a range of mitotic figure morphologies in an entire tissue sample.

Joseph R. Daniele, PhD, Institute Research Scientist

Dr. Daniele received his Ph.D. in Biochemistry from Harvard, focusing on protein trafficking and axonal transport of the oncogenic mediator Hedgehog. He then proceeded to Andrew Dillin’s lab at UC-Berkeley where his research focused on high-content analysis and characterization of the unfolded protein response.

Intrahepatic macrophages influence the composition of the microenvironment, host immune response to liver injury and development of fibrosis. In this webinar, Heather Stevenson will present her group’s findings from an analysis of five different antibodies commonly observed on macrophage populations (CD68, MAC387, CD163, CD14 and CD16).

Using a multiplex protocol, the group stained biopsies collected from representative patients with chronic liver diseases, including chronic hepatitis C, non-alcoholic steatohepatitis and autoimmune hepatitis. Spectral imaging microscopy and deep-learning-based analysis was applied and found to be a powerful tool that enables in situ analysis of macrophages and other cells in human liver biopsies and may lead to more personalized therapeutic approaches in the future.

• • How multispectral imaging and deep-learning-based image analysis facilitates comparison and visualization of macrophage populations

• • About spectral unmixing of fluorophore signals, subtraction of auto-fluorescence and preservation of hepatic architecture

• • How cell phenotyping, tissue segmentation and t-distributed stochastic neighbor embedding plots can facilitate characterization of numerous cell populations

• • How to optimize multiplex staining and spectral imaging microscopy

• • Discuss types of imaging analysis available for multiplex stained tissues

Heather Stevenson-Lerner, MD, Ph.D., FCAP, Asst. Professor, Dept. of Pathology Liver & Transplantation Pathologist, The University of Texas Medical Branch

Dr. Heather Stevenson-Lerner’s clinical focus includes liver, transplantation, and gastrointestinal pathology. Dr. Stevenson-Lerner completed a fellowship in the Transplant Pathology Division at the University of Pittsburgh Medical Center. Dr. Stevenson-Lerner leads UTMB’s Liver Diseases Diagnostic Management Team, which is popular with hepatologists, transplant coordinators, fellows, and residents.