Search

BackgroundAn increasing number of studies within digital pathology show the potential of artificial intelligence (AI) to diagnose cancer using histological whole slide images, which requires large and diverse data sets. While diversification may result in more generalizable AI-based systems, it can also introduce hidden variables. If neural networks are able to distinguish/learn hidden variables, these variables can introduce batch effects that compromise the accuracy of classification systems. ObjectiveThe objective of the study was to analyze the learnability of an exemplary selection of hidden variables (patient age, slide preparation date, slide origin, and scanner type) that are commonly found in whole slide image data sets in digital pathology and could create batch effects. MethodsWe trained four separate convolutional neural networks (CNNs) to learn four variables using a data set of digitized whole slide melanoma images from five different institutes. For robustness, each CNN training and evaluation run was repeated multiple times, and a variable was only considered learnable if the lower bound of the 95% confidence interval of its mean balanced accuracy was above 50.0%. ResultsA mean balanced accuracy above 50.0% was achieved for all four tasks, even when considering the lower bound of the 95% confidence interval. Performance between tasks showed wide variation, ranging from 56.1% (slide preparation date) to 100% (slide origin). ConclusionsBecause all of the analyzed hidden variables are learnable, they have the potential to create batch effects in dermatopathology data sets, which negatively affect AI-based classification systems. Practitioners should be aware of these and similar pitfalls when developing and evaluating such systems and address these and potentially other batch effect variables in their data sets through sufficient data set stratification.

Molecular hydrogen (H2) has antioxidant, anti-inflammatory, and anti-fibrotic effects. In a rat model simulating pulmonary fibrotic changes induced by monocrotaline-induced pulmonary hypertension (MPH), we had previously explored the impact of inhaled H2 on lung inflammation and blood pressure. In this study, we further focused the biological effects of H2 on mast cells (MCs) and the parameters of the fibrotic phenotype of the local tissue microenvironment. MPH resulted in a significantly increased number of MCs in both the pneumatic and respiratory parts of the lungs, an increased number of tryptase-positive MCs with increased expression of TGF-β, activated interaction with immunocompetent cells (macrophages and plasma cells) and fibroblasts, and increased MC colocalization with a fibrous component of the extracellular matrix of connective tissue. The alteration in the properties of the MC population occurred together with intensified collagen fibrillogenesis and an increase in the integral volume of collagen and elastic fibers of the extracellular matrix of the pulmonary connective tissue. The exposure of H2 together with monocrotaline (MCT), despite individual differences between animals, tended to decrease the intrapulmonary MC population and the severity of the fibrotic phenotype of the local tissue microenvironment compared to changes in animals exposed to the MCT effect alone. In addition, the activity of collagen fibrillogenesis associated with MCs and the expression of TGF-β and tryptase in MCs decreased, accompanied by a reduction in the absolute and relative content of reticular and elastic fibers in the lung stroma. Thus, with MCT exposure, inhaled H2 has antifibrotic effects involving MCs in the lungs of rats. This reveals the unknown development mechanisms of the biological effects of H2 on the remodeling features of the extracellular matrix under inflammatory background conditions of the tissue microenvironment. [ABSTRACT FROM AUTHOR]

Marfan syndrome (MFS) is a hereditary condition accompanied by disorders in the structural and regulatory properties of connective tissue, including elastic fibers, due to a mutation in the gene encodes for fibrillin-1 protein (FBN1 gene) and the synthesis of abnormal fibrillin-1 glycoprotein. Despite the high potential of mast cells (MCs) to remodel the extracellular matrix (ECM), their pathogenetic significance in MFS has not been considered yet. The group of patients with Marfan syndrome included two mothers and five children (three girls aged 4, 11, and 11 and two boys aged 12 and 13). Normal skin was examined in two children aged 11 and 12. Histochemical, monoplex, and multiplex immunohistochemical techniques; combined protocols of simultaneous histochemical and immunohistochemical staining (the results of staining were assessed using light, epifluorescence, and confocal microscopy); and bioinformatics algorithms for the quantitative analysis of detected targets were used to evaluate mast cells and their relationship with other cells from extracellular structures in the skin dermis. Analysis of the skin MC population in children with Marfan syndrome revealed a considerably increased number of intra-organic populations with the preservation of the specific Tryptase+Chymase+CPA3+ protease profile typical of the skin. The features of the MC histotopography phenotype in MFS consisted of closer colocalization with elastic fibers, smooth muscle cells, and fibroblasts. MCs formed many intradermal clusters that synchronized the activity of cell functions in the stromal landscape of the tissue microenvironment with the help of spatial architectonics, including the formation of cell chains and the creation of fibrous niches. In MCs, the expression of specific proteases, TGF-β, and heparin increased, with targeted secretion of biologically active substances relative to the dermal elastic fibers, which had specific structural features in MFS, including abnormal variability in thickness along their entire length, alternating thickened and thinned areas, and uneven surface topography. This paper discusses the potential role of MCs in strain analysis (tensometry) of the tissue microenvironment in MFS. Thus, the quantitative and qualitative rearrangements of the skin MC population in MFS are aimed at altering the stromal landscape of the connective tissue. The results obtained should be taken into account when managing clinical signs of MFS manifested in other pathogenetically critical structures of internal organs, including the aorta, tendons, cartilage, and parenchymal organs. [ABSTRACT FROM AUTHOR]

The mechanisms of the pathogenesis of neck paraganglioma (PGL) and the possible role of mast cells (MCs) in its development and metastasis are still poorly understood. We analyzed MCs' morphologic characterization, activation, and the properties of their cytoplasmic/released granules in PGLs, using light and transmission electron microscopy. Paragangliomas showed a large tumor‐associated MC population both in the connective tissue layers of the tumor and between the tumor cells. Notably, MCs were presented by a high expression of specific proteases, size variation, polymorphism, and variable ultrastructural phenotype of granules. A massive number of granules were released surrounding the degranulated MCs while the integrity of MC membrane was maintained. Granules were electron‐dense with or without a membrane, ranging from 0.2 to 0.8 μm in diameter. MC plasmalemma was not found at the site of MC‐collagen fibrils contact, whereas the secretome and fibrils were directly contacted. We observed direct and mediator‐based interactions between MCs and paraganglioma cells. The latter preserved their membrane integrity when MC granules were not in proximity. The effects of the MC secretome on the paraganglioma microenvironment demonstrated its pathogenetic role in tumor progression and allow its application to new diagnostic criteria and the development of protocols for personalized therapy. Research Highlights: Ultrastructural analysis reveals novel regulatory effects of mast cells via diverse secretory pathways on the pathogenesis of parasympathetic paraganglioma, including fibrous extracellular matrix remodeling and mediator‐based interactions between MCs and cells of the tumor microenvironment. [ABSTRACT FROM AUTHOR]

Epidermal growth factor receptor (EGFR) tyrosine kinase inhibitors (TKIs) represent first-line standard of care in unresectable EGFR mutation-positive (EGFRm+) non-small cell lung cancer (NSCLC). However, 10-20% of patients with EGFRm+ NSCLC have uncommon EGFR variants, defined as mutations other than L858R substitutions or exon 19 deletions. NSCLC harboring uncommon EGFR mutations may demonstrate lower sensitivity to targeted agents than NSCLC with L858R or exon 19 deletion mutations. Prospective clinical trial data in patients with NSCLC uncommon EGFR mutations are lacking. Afatinib is a second-generation TKI and the only Food and Drug Administration-approved drug for some of the more prevalent uncommon EGFR mutations. We present a series of seven case reports describing clinical outcomes in afatinib-treated patients with NSCLC harboring a diverse range of extremely rare mutations with or without co-mutations affecting other genes. EGFR alterations included compound mutations, P-loop αC-helix compressing mutations, and novel substitution mutations. We also present a case with NSCLC harboring a novel EGFR::CCDC6 gene fusion. Overall, the patients responded well to afatinib, including radiologic partial responses in six patients during treatment. Responses were durable for three patients. The cases presented are in line with a growing body of clinical and preclinical evidence that indicating that NSCLC with various uncommon EGFR mutations, with or without co-mutations, may be sensitive to afatinib. [ABSTRACT FROM AUTHOR]

Eosinophilic esophagitis (EoE) is an immune-mediated disease that manifests with dysphagia and is characterized by the predominantly eosinophilic infiltration of the esophageal mucosa. Several instruments have been developed to assess the symptoms of EoE: the Daily Symptom Questionnaire (DSQ), EoE Activity Index (EEsAI), Pediatric EoE Symptom Severity (PEESSv2), etc. The use of the EREFS is a gold standard for endoscopic diagnosis. The EoE histologic scoring system (EoEHSS) was elaborated for the assessment of histological features in EoE. However, the remission criteria are not clearly defined and vary greatly in different studies. Gastroenterologists establish the severity of EoE mainly based on endoscopic findings. At the same time, EoE requires a multidisciplinary approach. The recently developed Index of Severity of Eosinophilic Esophagitis (I-SEE) that is built on symptoms, endoscopic findings, and histological features is promising. [ABSTRACT FROM AUTHOR]