Explore the vast range of titles available.

The advent of whole-slide imaging in digital pathology has brought about the advancement of computer-aided examination of tissue via digital image analysis. Digitized slides can now be easily annotated and analyzed via a variety of algorithms. This study reviews the fundamentals of tissue image analysis and aims to provide pathologists with basic information regarding the features, applications, and general workflow of these new tools. The review gives an overview of the basic categories of software solutions available, potential analysis strategies, technical considerations, and general algorithm readouts. Advantages and limitations of tissue image analysis are discussed, and emerging concepts, such as artificial intelligence and machine learning, are introduced. Finally, examples of how digital image analysis tools are currently being used in diagnostic laboratories, translational research, and drug development are discussed.

BackgroundUnderstanding the complex interactions between tumor, immune, and stromal cells in the tumor microenvironment (TME) is essential for understanding how patients respond to treatments as well as for the development of next generation immunotherapies. Here, we demonstrate the multiomic capability of the RNAscope™ technology by combining it with protein detection to interrogate the TME of breast cancer and lung cancer tissue.MethodsA section of a tissue microarray containing breast and lung tissue was stained using RNAscope™ Multiplex V2 assay with target probes for PanCK, PD1, and CTLA4 and was stained using immunofluorescence (IF) for the PD-L1 checkpoint biomarker using the Integrated Co-Detection workflow from ACD. A second section was stained using RNAscope probes for TNFA, CCR5, and IFNG and was stained for CD4 using IF. Slides were scanned on Vectra Polaris scanner and were imported into the HALO® platform as QPTIFF images for image analysis. HALO AI was used to create an artifact classifier that removed highly autofluorescent red blood cells and blood vessels from the analysis. HALO AI was additionally used to create a tumor/stroma classifier to classify breast and lung tissue. The FISH IF module of HALO® was used to quantify RNA probes and IF signal in the tumor and the stroma compartments. Infiltration analysis using the HALO® Spatial Analysis module was performed on phenotypes of interest.ResultsPhenotypes of interest, such as PD-L1+ CTLA4+ (immune cells), PD-L1+ PanCK+ (tumor cells), PanCK+ (tumor cells), CD4+ TNFA+ IFNG+ (Th1 cells), and CD4+ CCR5+ (T cells) were analyzed across the breast and lung tumor and stromal compartments. In both tissue types, the percentage of CD4+ IFNG+ TNFA+ Th1 cells as well as the percentage of CD4+ CCR5+ T cells was greater in the stroma compared to the tumor compartment. FISH scores for each probe were calculated as well as a H-score for each probe in the four tissue compartments. An infiltration analysis of Th1 cells revealed distinct profiles in the stroma of lung and breast tumor tissue.ConclusionsWe demonstrate a powerful workflow to enable characterization and quantification of RNA and proteins expression in a single assay. This combined staining technique conserves tissue samples, provides flexibility to examine RNA and protein expression in the same sample, and enables detection of immune cell populations using antibodies and activation signatures using RNA probes against cytokines and chemokines.

Cellular subpopulations within the colorectal tumor microenvironment (TME) include CD3 + and CD8 + lymphocytes, CD68 + and CD163 + macrophages, and tumor buds (TBs), all of which have known prognostic significance in stage II colorectal cancer. However, the prognostic relevance of their spatial interactions remains unknown. Here, by applying automated image analysis and machine learning approaches, we evaluate the prognostic significance of these cellular subpopulations and their spatial interactions. Resultant data, from a training cohort retrospectively collated from Edinburgh, UK hospitals ( n  = 113), were used to create a combinatorial prognostic model, which identified a subpopulation of patients who exhibit 100% survival over a 5-year follow-up period. The combinatorial model integrated lymphocytic infiltration, the number of lymphocytes within 50-μm proximity to TBs, and the CD68 + /CD163 + macrophage ratio. This finding was confirmed on an independent validation cohort, which included patients treated in Japan and Scotland ( n  = 117). This work shows that by analyzing multiple cellular subpopulations from the complex TME, it is possible to identify patients for whom surgical resection alone may be curative.

BackgroundInflammatory tumor micro-environments contain cells of various types and sub-types. The composition and spatial location of the cell populations reflects the host reaction to the inflammatory stimulus and increasingly is understood to influence responsiveness to tumor immunotherapies. Multiplexed imaging technologies allow identification of cell types and states within the spatial context of tissue architecture. We present here a prototype workflow that combines rapid high-resolution, whole-slide highly multiplexed immunofluorescence imaging with advanced image analysis tools for 1) segmenting tissues, cells, and quantifying cellular phenotypes based on multiple markers and 2) determining regional densities and proximity between cells. We apply the workflow to comparative assessment of three lymphoid tissues: tonsil (follicular hyperplasia); lymph node (quiescence); lymphoma (architectural effacement).MethodsFormalin-fixed, paraffin-embedded 5 micron sections of tonsil, lymph node and chronic lymphocytic leukemia/small lymphocytic lymphoma were deparaffinized, subjected to alkaline pH epitope retrieval, and then manually stained with a 17-plex panel including CD45 (leukocytes); CD20 (B cells); CD3d, CD4, CD8 (T cells); FOXP3 (T reg cells); CD68, CD163 (macrophages); CD45RO (activated cells); PD-L1, PD-1 (checkpoint markers); CD31 (vascular and lymphatic endothelial cells); cytokeratin, E-cadherin (epithelial cells); PCNA, Ki-67 (proliferating cells); and a nuclear dye. Stained slides were coverslipped and imaged on the Orion Instrument (RareCyte) generating .ome tiff image files. The HighPlex FL module of the HALO image analysis platform from Indica Labs with embedded HALO AI performed nuclear and cell segmentation, nuclear phenotyping, and user-defined thresholds were applied to each of the biomarkers to define positivity for the appropriate subcellular localization (nuclear, cytoplasmic, and/or membrane) for phenotypic analysis. H & E images from either the same or serial sections were integrated with the multiplex images using the HALO Serial Stain module.ResultsRegional masks that were defined by predominance of B-cells (CD20) or T-cells (CD3d) matched known lymphoid micro-anatomy of follicles and inter-follicular cortex respectively. Within the regions, populations and sub-populations of B-cells, T-cells, macrophages and vessels were measured, and their densities calculated and compared between tissues. Rare cell types of potential importance in immuno-oncology were investigated. The results demonstrate differences between the tissues at a phenotypic level that correspond to the morphologic differences seen by light microscopy.ConclusionsOrion imaging combined with HALO image analysis provides a powerful and intuitive workflow for visualization and quantification of distinct microenvironment populations for use in translational and clinical research.

The categorisation of desmoplastic reaction (DR) present at the colorectal cancer (CRC) invasive front into mature, intermediate or immature type has been previously shown to have high prognostic significance. However, the lack of an objective and reproducible assessment methodology for the assessment of DR has been a major hurdle to its clinical translation. In this study, a deep learning algorithm was trained to automatically classify immature DR on haematoxylin and eosin digitised slides of stage II and III CRC cases (n = 41). When assessing the classifier’s performance on a test set of patient samples (n = 40), a Dice score of 0.87 for the segmentation of myxoid stroma was reported. The classifier was then applied to the full cohort of 528 stage II and III CRC cases, which was then divided into a training (n = 396) and a test set (n = 132). Automatically classed DR was shown to have superior prognostic significance over the manually classed DR in both the training and test cohorts. The findings demonstrated that deep learning algorithms could be applied to assist pathologists in the detection and classification of DR in CRC in an objective, standardised and reproducible manner.

Cellular subpopulations within the colorectal tumor microenvironment (TME) include CD3 and CD8 lymphocytes, CD68 and CD163 macrophages, and tumor buds (TBs), all of which have known prognostic significance in stage II colorectal cancer. However, the prognostic relevance of their spatial interactions remains unknown. Here, by applying automated image analysis and machine learning approaches, we evaluate the prognostic significance of these cellular subpopulations and their spatial interactions. Resultant data, from a training cohort retrospectively collated from Edinburgh, UK hospitals (n = 113), were used to create a combinatorial prognostic model, which identified a subpopulation of patients who exhibit 100% survival over a 5-year follow-up period. The combinatorial model integrated lymphocytic infiltration, the number of lymphocytes within 50-μm proximity to TBs, and the CD68 /CD163 macrophage ratio. This finding was confirmed on an independent validation cohort, which included patients treated in Japan and Scotland (n = 117). This work shows that by analyzing multiple cellular subpopulations from the complex TME, it is possible to identify patients for whom surgical resection alone may be curative.

Somatic cells from persons with the inherited chromosome breakage syndrome Bloom syndrome (BS) feature excessive chromosome breakage, intra- and inter-chromosomal homologous exchanges and telomeric associations. The gene mutated in BS, BLM, encodes a RecQ-like ATP-dependent 3′-to-5 ′ helicase that presumably functions in some types of DNA transactions. As the absence of BLM is associated with excessive recombination, in vitro experiments have tested the ability of BLM to suppress recombination and/or resolve recombination intermediates. In vitro, BLM promotes branch migration of Holliday junctions, resolves D-loops and unwinds G-quadruplex DNA. A function for BLM in maintaining telomeres is suggested by the latter, since D-loops and perhaps G-quadruplex structures are thought to be present at telomeres. In the present study, the association of BLM with telomeres was investigated. Given the association of BLM with recombination, it was of particular interest to determine the nuclear localization of BLM with respect to telomeres in cells using recombinational pathways for telomere lengthening, termed ALT. Using the telomere repeat protein TRF2 as a telomere marker, we demonstrate that BLM co-localizes with telomeres in cells using ALT, but not in telomerase-positive or primary cells. BLM colocalizes with TRF2 in foci actively synthesizing DNA during late S and G2/M; colocalization is enriched during these phases of the cell cycle when ALT is thought to occur. By immunoprecipitation, BLM associates with telomeres and TRF2 in cells using ALT. In S. cerevisiae , we demonstrate that BLM expression rescues a defect in recombinational telomere lengthening associated with absence of SGS1. These data establish a spatial and temporal association of BLM with telomere synthesis in cells using ALT and demonstrate conserved function(s) for BLM and SGS1 in ALT. Additionally, the regulation of BLM activity using telomere substrates was investigated in vitro. We find that TRF1 and TRF2 physically and functionally interact with BLM in vitro. TRF2 stimulates BLM unwinding of telomeric and non-telomeric substrates. Conversely, TRF1 inhibits BLM unwinding of telomeric substrates only. Neither TRF1 nor TRF2 regulate unwinding activity of the UvrD helicase. Finally, BLM helicase activity is stimulated by TRF2 with equimolar concentrations of TRF1, but not when TRF1 is present in molar excess. Based on these data, we present a model for the coordinated regulation of BLM helicase activity by TRF1 and TRF2 at telomeres in cells using ALT.