Standardized Classification of Lung Adenocarcinoma Subtypes and Improvement of Grading Assessment Through Deep Learning.

The histopathologic distinction of lung adenocarcinoma (LADC) subtypes is subject to high interobserver variability, which can compromise the optimal assessment of patient prognosis. Therefore, this study developed convolutional neural networks capable of distinguishing LADC subtypes and predicting disease-specific survival, according to the recently established LADC tumor grades. Consensus LADC histopathologic images were obtained from 17 expert pulmonary pathologists and one pathologist in training. Two deep learning models (AI-1 and AI-2) were trained to predict eight different LADC classes. Furthermore, the trained models were tested on an independent cohort of 133 patients. The models achieved high precision, recall, and F1 scores exceeding 0.90 for most of the LADC classes. Clear stratification of the three LADC grades was reached in predicting the disease-specific survival by the two models, with both Kaplan-Meier curves showing significance (P = 0.0017 and 0.0003). Moreover, both trained models showed high stability in the segmentation of each pair of predicted grades with low variation in the hazard ratio across 200 bootstrapped samples. These findings indicate that the trained convolutional neural networks improve the diagnostic accuracy of the pathologist and refine LADC grade assessment. Thus, the trained models are promising tools that may assist in the routine evaluation of LADC subtypes and grades in clinical practice.

Comparative study on a novel lobule structure of the zebrafish liver and that of the mammalian liver.

The mammalian liver has a lobule structure with a portal triad consisting of the portal vein, hepatic artery, and bile duct, which exhibits zonal gene expression, whereas those of teleosts do not have a portal triad. It remains to be demonstrated what kind of the unit structures they have, including their gene expression patterns. The aims of the present study were to demonstrate the unit structure of the teleost liver and discuss it in terms of evolution and adaptation in vertebrates and the use of teleosts as an alternative model for human disease. The zebrafish liver was examined as a representative of teleosts with respect to its morphological architecture and gene expression. A novel, polygonal lobule structure was detected in the zebrafish liver. In it, portal veins and central veins were distributed at the periphery and center, respectively. Sinusoids connected both veins. Anxa4-positive preductules were incorporated into the tubular lumen of two rows of hepatocytes in sections. Intrahepatic bile ducts resided randomly in the liver lobule. Zebrafish livers did not have zonal gene expression for metabolic pathways examined. The lobules of the zebrafish liver with preductules located in the tubular lumina of hepatocytes may resemble the oval cell reaction of injured livers of mammals and might convey bile to the intestine more safely than mammalian livers. The gene expression pattern in liver lobules and our liver lobule model of the zebrafish may be important to discuss data obtained in experiments using this animal as an alternative model for human disease.

Phylogenetic analyses of the hepatic architecture in vertebrates.

The mammalian liver has a structural and functional unit called the liver lobule, in the periphery of which the portal triad consisting of the portal vein, bile duct and hepatic artery is developed. This type of hepatic architecture is detectable in many other vertebrates, including amphibians and birds, whereas intrahepatic bile ducts run independently of portal vein distribution in actinopterygians such as the salmon and tilapia. It remains to be clarified how the hepatic architectures are phylogenetically developed among vertebrates. The present study morphologically and immunohistochemically analyzed the hepatic structures of various vertebrates, including as many classes and subclasses as possible, with reference to intrahepatic bile duct distribution. The livers of vertebrates belonging to the Agnatha, Chondrichthyes, Amphibia, Aves, Mammalia, and Actinopterygii before Elopomorpha, had the portal triad-type architecture. The Anguilliformes livers developed both periportal bile ducts and non-periportal bile ducts. The Otocephala and Euteleostei livers had independent configuration of bile ducts and portal veins. Pancreatic tissues penetrated the liver parenchyma along portal veins in the Euteleostei. The liver of the lungfish, which shares the same origin with amphibians, did not have the portal triad-type architecture. Teleostei and lungfish livers had ductular development in the liver parenchyma similar to oval cell proliferation in injured mammalian livers. Euteleostei livers had penetration of significant numbers of independent portal veins from their intestines, suggesting that each liver lobe might receive a different blood supply. The hepatic architectures of the portal triad-type changed to non-portal triad-type architecture along the evolution of the Actinopterygii. The hepatic architecture of the lungfish resembles that of the Actinopterygii after Elopomorpha in intrahepatic biliary configuration, which may be an example of convergent evolution.

Gene expression in the liver of the hagfish (Eptatretus burgeri) belonging to the Cyclostomata is ancestral to that of mammals.

Although the liver of the hagfish, an earliest diverged lineage among vertebrates, has a histological architecture similar to that of mammals, its gene expression has not been explored yet. The present study was undertaken to comparatively characterize gene expression in the liver of the hagfish with that of the mouse, using in situ hybridization technique. Expression of alb (albumin) was detectable in all hepatocytes of the hagfish liver, but was negative in intrahepatic bile ducts. Their expression in abundant periportal ductules was weak. The expression pattern basically resembled that in mammalian livers, indicating that the differential expression of hepatocyte markers in hepatocytes and biliary cells may have been acquired in ancestral vertebrates. alb expression was almost homogeneous in the hagfish liver, whereas that in the mouse liver lobule was zonal. The glul (glutamate-ammonia ligase) expression was also homogeneously detectable in hepatocytes without zonation, and weakly so in biliary cells of the hagfish, which contrasted with its restricted pericentral expression in mouse livers. These findings indicated that the hagfish liver did not have mammalian-type zonation. Whereas tetrapods had Hnf (hepatocyte nuclear factor) 1a and Hnf1b genes encoding the transcription factors, the hagfish had a single gene of their orthologue hnf1. Although HNF1α and HNF1ß were immunohistochemically detected in hepatocytes and biliary cells of the mouse, respectively, hnf1 was expressed in both hepatocytes and biliary cells of the hagfish. These data indicate that gene expression of hnf1 in the hagfish liver may be ancestral with that of alb and glul during vertebrate evolution.

Inversin-deficient (inv) mice do not establish a polarized duct system in the liver and pancreas.

Inversin-deficient (inv) mice have anomalies in liver and pancreatic development in addition to an inverted left-right axis of the body. The present study was undertaken to unveil mechanisms of bile and pancreatic duct development from immunohistochemical analyses of anomalies in inv mice. Intrahepatic bile ducts having proximodistal polarity in size and the height of their epithelia, and ductules were formed in livers of wild-type neonates. By contrast, in inv mice, ductal plates, precursor structures of intrahepatic bile ducts and ductules, persisted without the proximodistal polarity. Their epithelial cells did not acquire planar cell polarity (PCP) in terms of expression of tight junction proteins although they expressed bile duct markers, HNF1ß and SOX9. They had an apicobasal polarity from expression of basal laminar components. Enlargement of the hepatic artery and poor connective tissue development, including the abnormal deposition of the extracellular matrices, were also noted in inv mice, suggesting that bile duct development was coupled to that of the hepatic artery and portal vein. In pancreata of inv neonates, neither the main pancreatic duct was formed, nor dilated duct-like structures had the morphological polarity from the connecting point with the common bile duct. Lumina of acini was dilated, and centroacinar cells changed their position in the acini to their neck region. Immunohistochemical analyses of tight junction proteins suggested that epithelial cells of the duct-like structures did not have a PCP. Thus, Invs may be required for the establishment of the PCP of the whole duct system in the liver and pancreas.

Changes of biliary cilia, smooth muscle tissue distribution, innervation and extracellular matrices during morphological evolution of hepatic architectures in vertebrates.

BACKGROUND: The liver architecture of vertebrates can be classified into two types, the portal triad type (having periportal bile ducts) and the non-portal triad type (having bile ducts independent of the course of portal veins). The former is typically detectable in livers of tetrapods and cartilaginous fish, and its ancestral state is found in the hagfish, an earliest diverged lineage among vertebrates. Teleosts other than osteoglossomorphs have the latter. The aim of the present study is to reveal the changes of the hepatic innervation, biliary cilia and smooth muscle distribution, and extracellular matrices along vertebrate evolution with attention to the two types of liver architectures. METHODS: The hepatic innervation, biliary cilia and smooth muscle distribution, and collagen deposition were immunohistochemically and histochemically compared among livers of various vertebrates, using anti-acetylated tubulin and anti-α-smooth muscle actin antibodies, and Sirius red staining. These were also ultrastructurally examined. RESULTS: Although the hagfish liver had periportal intrahepatic bile ducts and ductules as detected in mammalian livers, it lacked smooth muscles around bile ducts and portal veins. Extracellular matrices in their connective tissues had thick collagen fibers. Its innervation was restricted to intrahepatic bile ducts and portal veins in the hilum. In livers of other vertebrates, including teleosts, the innervation was broadly detectable, especially around bile ducts, hepatic arteries and portal veins (afferent vessels), but not around central veins (efferent vessels). The chondrichthyans ultrastructurally had smooth muscle tissue around bile ducts. Cilia distribution was confirmed in intrahepatic bile ducts of tetrapods and basal actinopterygians. Teleosts other than osteoglossomorphs lacked cilia in their intrahepatic bile ducts. CONCLUSIONS: The liver architecture of the hagfish may be unique for innervation and extracellular matrices. Hepatic innervation may not have occurred in vertebrate ancestors. Hepatic innervation in bile ducts, hepatic arteries and portal veins may have been conserved among the extant jawed vertebrates. Cilia distribution in bile ducts may have changed during evolution of actinopterygians.

In vitro-based prediction of human plasma concentrations of food-related compounds.

Efforts have been made to replace animal experiments in safety evaluations, including in vitro-based predictions of human internal exposures, such as predicting peak plasma concentration (Cmax) values for xenobiotics and comparing these values with in vitro-based toxicity endpoints. Herein, the authors predicted the Cmax values of food-related compounds in humans based on existing and novel in vitro techniques. In this study, 20 food-related compounds, which have been previously reported in human pharmacokinetic or toxicokinetic studies, were evaluated. Human induced pluripotent stem cell-derived small intestinal epithelial cells (hiPSC-SIEC) and Caco-2 cells, HepaRG cells, equilibrium dialysis of human plasma, and LLC-PK1 cell monolayer were used to assess intestinal absorption and availability, hepatic metabolism, unbound plasma fraction, and secretion and reabsorption in renal tubular cells, respectively. After conversion of these parameters into human kinetic parameters, the plasma concentration profiles of these compounds were predicted using in silico methods, and the obtained Cmax values were found to be between 0.017 and 183 times the reported Cmax values. When the in silico-predicted parameters were modified with in vitro data, the predicted Cmax values came within 0.1-10 times the reported values because the metabolic activities of hiPSC-SIECs, such as uridine 5'-diphospho-glucuronosyl transferase, are more similar to those of human primary enterocytes. Thus, combining in vitro test results with the plasma concentration simulations resulted in more accurate and transparent predictions of Cmax values of food-related compounds than those obtained using in silico-derived predictions alone. This method facilitates accurate safety evaluation without the need for animal experiments.

Immunohistological analysis on distribution of smooth muscle tissues in livers of various vertebrates with attention to different liver architectures.

BACKGROUND: The liver architecture of vertebrates can be classified into two types, the portal triad type (having periportal bile ducts) and the non-portal triad type (having non-periportal bile ducts). The former is detectable from the hagfish, which is the most ancestral vertebrate, to tetrapod livers whereas many actinopterygian livers have the latter. The aim of the present study is to reveal the distribution of smooth muscle tissue in livers of various vertebrates with attention to their architectures. METHODS: Smooth muscle was immunohistochemically compared in hepatic blood vessels and bile ducts of various vertebrates, using an anti-alpha-smooth muscle actin (ASMA) antibody. RESULTS: Smooth muscle was noted in the gallbladder and hepatic artery in all vertebrates, including the hagfish. Bile ducts having ASMA-positive smooth muscles were absent in the hagfish, but detected in the Chondrichthyes and conserved in actinopterygians with or without portal triads during the evolution of vertebrates. In tetrapods having portal triads, reptiles had a tendency to have strongly ASMA-positive biliary smooth muscle tissues whereas other tetrapods had bile ducts with poor smooth muscle tissues. Although the hagfish livers never had ASMA-positive smooth muscle tissue in the walls of portal and central veins, it was observed in discontinuous distributions or not observed in portal veins and central veins of chondrichthyans and actinopterygians. By contrast, in most tetrapods, ASMA-positive smooth muscle tissue was detectable in portal veins, which supported the adjacent endothelial cells as a circular layer. Central veins did not consistently have smooth muscle tissue in these groups. DISCUSSION AND CONCLUSION: The hagfish liver may retain more ancestral characteristics than other vertebrates in terms of smooth muscle distribution in the vascular and biliary systems. Actinopterygians might have a different mechanism of bile transport from tetrapods from their smooth muscle distribution in intrahepatic bile ducts. The circular smooth muscle distribution in portal veins might be a characteristic acquired by tetrapods.