Explore the vast range of titles available.

For virtually every patient with colorectal cancer (CRC), hematoxylin-eosin (HE)-stained tissue slides are available. These images contain quantitative information, which is not routinely used to objectively extract prognostic biomarkers. In the present study, we investigated whether deep convolutional neural networks (CNNs) can extract prognosticators directly from these widely available images. We hand-delineated single-tissue regions in 86 CRC tissue slides, yielding more than 100,000 HE image patches, and used these to train a CNN by transfer learning, reaching a nine-class accuracy of >94% in an independent data set of 7,180 images from 25 CRC patients. With this tool, we performed automated tissue decomposition of representative multitissue HE images from 862 HE slides in 500 stage I-IV CRC patients in the The Cancer Genome Atlas (TCGA) cohort, a large international multicenter collection of CRC tissue. Based on the output neuron activations in the CNN, we calculated a \"deep stroma score,\" which was an independent prognostic factor for overall survival (OS) in a multivariable Cox proportional hazard model (hazard ratio [HR] with 95% confidence interval [CI]: 1.99 [1.27-3.12], p = 0.0028), while in the same cohort, manual quantification of stromal areas and a gene expression signature of cancer-associated fibroblasts (CAFs) were only prognostic in specific tumor stages. We validated these findings in an independent cohort of 409 stage I-IV CRC patients from the \"Darmkrebs: Chancen der Verhütung durch Screening\" (DACHS) study who were recruited between 2003 and 2007 in multiple institutions in Germany. Again, the score was an independent prognostic factor for OS (HR 1.63 [1.14-2.33], p = 0.008), CRC-specific OS (HR 2.29 [1.5-3.48], p = 0.0004), and relapse-free survival (RFS; HR 1.92 [1.34-2.76], p = 0.0004). A prospective validation is required before this biomarker can be implemented in clinical workflows. In our retrospective study, we show that a CNN can assess the human tumor microenvironment and predict prognosis directly from histopathological images.

The thymus prevents autoimmune diseases through mechanisms that operate in the cortex and medulla, comprising positive and negative selection and the generation of regulatory T-cells (Tregs). Egress from the thymus through the perivascular space (PVS) to the blood is another possible checkpoint, as shown by some autoimmune/immunodeficiency syndromes. In polygenic autoimmune diseases, subtle thymic dysfunctions may compound genetic, hormonal and environmental cues. Here, we cover (a) tolerance-inducing cell types, whether thymic epithelial or tuft cells, or dendritic, B- or thymic myoid cells; (b) tolerance-inducing mechanisms and their failure in relation to thymic anatomic compartments, and with special emphasis on human monogenic and polygenic autoimmune diseases and the related thymic pathologies, if known; (c) polymorphisms and mutations of tolerance-related genes with an impact on positive selection (e.g. the gene encoding the thymoproteasome-specific subunit, PSMB11), promiscuous gene expression (e.g. AIRE, PRKDC, FEZF2, CHD4), Treg development (e.g. SATB1, FOXP3), T-cell migration (e.g. TAGAP) and egress from the thymus (e.g. MTS1, CORO1A); (d) myasthenia gravis as the prototypic outcome of an inflamed or disordered neoplastic ‘sick thymus’.

Automatic recognition of different tissue types in histological images is an essential part in the digital pathology toolbox. Texture analysis is commonly used to address this problem; mainly in the context of estimating the tumour/stroma ratio on histological samples. However, although histological images typically contain more than two tissue types, only few studies have addressed the multi-class problem. For colorectal cancer, one of the most prevalent tumour types, there are in fact no published results on multiclass texture separation. In this paper we present a new dataset of 5,000 histological images of human colorectal cancer including eight different types of tissue. We used this set to assess the classification performance of a wide range of texture descriptors and classifiers. As a result, we found an optimal classification strategy that markedly outperformed traditional methods, improving the state of the art for tumour-stroma separation from 96.9% to 98.6% accuracy and setting a new standard for multiclass tissue separation (87.4% accuracy for eight classes). We make our dataset of histological images publicly available under a Creative Commons license and encourage other researchers to use it as a benchmark for their studies.

Lymphoid and myeloid cells are abundant in the tumor microenvironment, can be quantified by immunohistochemistry and shape the disease course of human solid tumors. Yet, there is no comprehensive understanding of spatial immune infiltration patterns (‘topography’) across cancer entities and across various immune cell types. In this study, we systematically measure the topography of multiple immune cell types in 965 histological tissue slides from N = 177 patients in a pan-cancer cohort. We provide a definition of inflamed (‘hot’), non-inflamed (‘cold’) and immune excluded patterns and investigate how these patterns differ between immune cell types and between cancer types. In an independent cohort of N = 287 colorectal cancer patients, we show that hot, cold and excluded topographies for effector lymphocytes (CD8) and tumor-associated macrophages (CD163) alone are not prognostic, but that a bivariate classification system can stratify patients. Our study adds evidence to consider immune topographies as biomarkers for patients with solid tumors.

Background Histological images show strong variance (e.g. illumination, color, staining quality) due to differences in image acquisition, tissue processing, staining, etc. This can impede downstream image analysis such as staining intensity evaluation or classification. Methods to reduce these variances are called image normalization techniques. Methods In this paper, we investigate the potential of CycleGAN (cycle consistent Generative Adversarial Network) for color normalization in hematoxylin-eosin stained histological images using daily clinical data with consideration of the variability of internal staining protocol variations. The network consists of a generator network G B that learns to map an image X from a source domain A to a target domain B , i.e. G B : X A → X B . In addition, a discriminator network D B is trained to distinguish whether an image from domain B is real or generated. The same process is applied to another generator-discriminator pair ( G A , D A ), for the inverse mapping G A : X B → X A . Cycle consistency ensures that a generated image is close to its original when being mapped backwards ( G A ( G B ( X A ))≈ X A and vice versa). We validate the CycleGAN approach on a breast cancer challenge and a follicular thyroid carcinoma data set for various stain variations. We evaluate the quality of the generated images compared to the original images using similarity measures. In addition, we apply stain normalization on pathological lymph node data from our institute and test the gain from normalization on a ResNet classifier pre-trained on the Camelyon16 data set. Results Qualitative results of the images generated by our network are compared to original color distributions. Our evaluation indicates that by mapping images to a target domain, the similarity training images from that domain improves up to 96%. We also achieve a high cycle consistency for the generator networks by obtaining similarity indices greater than 0.9. When applying the CycleGAN normalization to HE-stain images from our institute the kappa-value of the ResNet-model that is only trained on Camelyon16 data is increased more than 50%. Conclusions CycleGANs have proven to efficiently normalize HE-stained images. The approach compensates for deviations resulting from image acquisition (e.g. different scanning devices) as well as from tissue staining (e.g. different staining protocols), and thus overcomes the staining variations in images from various institutions.The code is publicly available at https://github.com/m4ln/stainTransfer_CycleGAN_pytorch . The data set supporting the solutions is available at https://doi.org/10.11588/data/8LKEZF .

Background Compilation of different morphological lesion signatures is characteristic of renal pathology. Previous studies have documented the potential value of artificial intelligence (AI) in recognizing relatively clear-cut glomerular structures and patterns, such as segmental or global sclerosis or mesangial hypercellularity. This study aimed to test the capacity of deep learning algorithms to recognize complex glomerular structural changes that reflect common diagnostic dilemmas in nephropathology. Methods For this purpose, we defined nine classes of glomerular morphological patterns and trained twelve convolutional neuronal network (CNN) models on these. The two-step training process was done on a first dataset defined by an expert nephropathologist (12,253 images) and a second consensus dataset (11,142 images) defined by three experts in the field. Results The efficacy of CNN training was evaluated using another set with 180 consensus images, showing convincingly good classification results (kappa-values 0.838–0.938). Furthermore, we elucidated the image areas decisive for CNN-based decision making by class activation maps. Finally, we demonstrated that the algorithm could decipher glomerular disease patterns coinciding in a single glomerulus (e.g. necrosis along with mesangial and endocapillary hypercellularity). Conclusions In summary, our model, focusing on glomerular lesions detectable by conventional microscopy, is the first sui generis to deploy deep learning as a reliable and promising tool in recognition of even discrete and/or overlapping morphological changes. Our results provide a stimulus for ongoing projects that integrate further input levels next to morphology (such as immunohistochemistry, electron microscopy, and clinical information) to develop a novel tool applicable for routine diagnostic nephropathology.

Acute hypoxemic respiratory failure is a critical challenge in intensive care. A substantial proportion of patients present with asymmetric acute lung injury (ALI), complicating management due to heterogeneous lung involvement. While lung-protective mechanical ventilation represents the standard of care, the optimal approach to positive end-expiratory pressure (PEEP) titration remains unclear. This study investigated the effects of transpulmonary pressure (TPP)-guided PEEP titration vs. a fixed PEEP strategy in a porcine model of unilateral ALI. A total of 14 pigs underwent ALI induction via unilateral surfactant depletion and were randomized to receive either a fixed PEEP of 5 cmH2O or a PEEP targeting a slightly positive TPP at end-expiration. Over six hours, respiratory mechanics, high-resolution computed tomography (HRCT), histological lung injury scores (LIS), and plasma protein biomarkers were assessed. TPP-guided PEEP titration significantly lowered driving pressure and improved compliance compared to fixed low PEEP, suggesting more homogeneous tidal volume distribution. HRCT revealed less collateral injury in the initially non-injured lung in the TPP-guided group. However, histopathological LIS did not differ between groups. Exploratory cytokine profiling showed systemic inflammatory activation—including pro- and anti-inflammatory responses—only in the TPP-guided group. These findings indicate that TPP-guided PEEP titration may optimize ventilation by balancing alveolar recruitment and overdistension in asymmetric ALI, with clear effects on physiological and imaging parameters, but without parallel effects on cytokine responses. Further research is needed to assess its long-term impact and clinical relevance.

A unique feature of thymomas is their unrivaled frequency of associated myasthenia gravis (MG). Previous studies reported that MG+ thymomas contain a larger number of mature \"pre-emigrant\" CD4+ T cells than MG- thymomas and that most thymomas do not contain AIRE expressing cells irrespective of MG status. These findings suggest that CD4+ T cells that mature inside the abnormal microenvironment of thymomas and egress to the blood are critical to the development of thymoma-associated MG (TAMG) irrespective of thymoma histotype. However, underlying mechanisms have remained enigmatic. To get hints to mechanisms underlying TAMG, we pursue three hypotheses: (i) Functional pathways with metabolic and immunological relevance might be differentially expressed in TAMG(+) compared to TAMG(-) thymomas; (ii) differentially enriched pathways might be more evident in immature lymphocyte-poor (i.e., tumor cell/stroma-rich) thymoma subgroups; and (iii) mechanisms leading to TAMG might be different among thymoma histological subtypes. To test these hypotheses, we compared the expression of functional pathways with potential immunological relevance ( = 380) in relation to MG status separately in type AB and B2 thymomas and immature lymphocyte-rich and lymphocyte-poor subgroups of these thymoma types using the TCGA data set. We found that <10% of the investigated pathways were differentially upregulated or downregulated in MG+ compared to MG- thymomas with significant differences between AB and B2 thymomas. The differences were particularly evident, when epithelial cell/stroma-rich subsets of type AB and B2 thymomas were analyzed. Unexpectedly, some MG-associated pathways that were significantly upregulated in AB thymomas were significantly downregulated in B2 thymomas, as exemplified by the oxidative phosphorylation pathway. Conversely, the MG-associated pathway related to macrophage polarization was downregulated in MG+ AB thymoma and upregulated in MG+ B2 thymoma. We conclude that functional pathways are significantly associated with TAMG, and that some mechanisms leading to TAMG might be different among thymoma histological subtypes. Functions related to metabolisms, vascular and macrophage biology are promising new candidate mechanisms potentially involved in the pathogenesis of TAMG. More generally, the results imply that future studies addressing pathomechanisms of TAMG should take the histotype and abundance of tumor cells and non-neoplastic stromal components of thymomas into account.

Thymomas exhibit a unique genomic landscape, comprising the lowest on average total mutational burden among adult human cancers; a unique point mutation in the GTF2I gene in WHO type A and AB thymomas (and rarely others); almost unique KMT2A-MAML2 translocations in rare WHO type B2 and B3 thymomas; a unique YAP1-MAML2 translocation in almost all metaplastic thymomas; and unique miRNA profiles in relation to GTF2I mutational status and WHO histotypes. While most thymomas can be diagnosed solely on the basis of morphological features, mutational analyses can solve challenging differential diagnostic problems. No molecular biomarkers have been identified that predict the response of unresectable thymomas to chemotherapy or agents with known molecular targets. Despite the common and strong expression of PDL1 in thymomas, immune checkpoint inhibitors are rarely applicable due to the poor predictability of common, life-threatening autoimmune side effects that are related to the unrivaled propensity of thymomas towards autoimmunity.

Renal cell carcinoma (RCC) is the deadliest primary genitourinary malignancy typically associated with asymptomatic initial presentation and poorly predictable survival. Next to established risk factors, tumor microenvironment may alter metastatic capacity and immune landscape. Due to their high concentrations, sulfoglycolipids (sulfatides) were among the first well-described antigens in RCC that are associated with worse prognosis. As sulfatide detection in routine diagnostics is not possible, we aimed to test the prognostic value of its protein counterpart, sulfatide-producing enzyme Gal3ST1. We performed retrospective long-term follow up analysis of Gal3ST1 expression as prognostic risk factor in a representative RCC patient cohort. We observed differentially regulated Gal3ST1 expression in all RCC types, being significantly more associated with clear cell RCC than to chromophobe RCC (p = 0.001). Surprisingly, in contrast to published observations from in vitro models, we could not confirm an association between Gal3ST1 expression and a malignant clinical behaviour of the RCC. In our cohort, Gal3ST1 did not significantly influence progression-free survival (Hazard Ratio (HR): 1.7 95% CI (0.6–4.9), p = 0.327). Particularly after adjusting for histology, T-stage, N-status and M-status at baseline, we observed no independent prognostic effect (HR = 1.0 95% CI (0.3–3.3), p = 0.96). The analysis of Gal3ST1 mRNA expression in a TCGA dataset supported the results of our cohort. Thus, Gal3ST1 might help to differentiate between chromophobe RCC and other frequent RCC entities but—despite previously published data from cell culture models—does not qualify as a prognostic marker for RCC. Further investigation of regulatory mechanisms of sulfatide metabolism in human RCC microenvironment is necessary to understand the role of this quantitatively prominent glycosphingolipid in RCC progression.