Search

Background & Aims: Diagnosis of adenocarcinoma in the liver is a frequent scenario in routine pathology and has a critical impact on clinical decision making. However, rendering a correct diagnosis can be challenging, and often requires the integration of clinical, radiologic, and immunohistochemical information. We present a deep learning model (HEPNET) to distinguish intrahepatic cholangiocarcinoma from colorectal liver metastasis, as the most frequent primary and secondary forms of liver adenocarcinoma, with clinical grade accuracy using H&E-stained whole-slide images., Methods: HEPNET was trained on 714,589 image tiles from 456 patients who were randomly selected in a stratified manner from a pool of 571 patients who underwent surgical resection or biopsy at Heidelberg University Hospital. Model performance was evaluated on a hold-out internal test set comprising 115 patients and externally validated on 159 patients recruited at Mainz University Hospital., Results: On the hold-out internal test set, HEPNET achieved an area under the receiver operating characteristic curve of 0.994 (95% CI, 0.989-1.000) and an accuracy of 96.522% (95% CI, 94.521%-98.694%) at the patient level. Validation on the external test set yielded an area under the receiver operating characteristic curve of 0.997 (95% CI, 0.995-1.000), corresponding to an accuracy of 98.113% (95% CI, 96.907%-100.000%). HEPNET surpassed the performance of 6 pathology experts with different levels of experience in a reader study of 50 patients (P = .0005), boosted the performance of resident pathologists to the level of senior pathologists, and reduced potential downstream analyses., Conclusions: We provided a ready-to-use tool with clinical grade performance that may facilitate routine pathology by rendering a definitive diagnosis and guiding ancillary testing. The incorporation of HEPNET into pathology laboratories may optimize the diagnostic workflow, complemented by test-related labor and cost savings., (Copyright © 2023 AGA Institute. Published by Elsevier Inc. All rights reserved.)

Background & Aims: Cholangiocyte proliferation and ductular reaction contribute to the onset and progression of liver diseases. Little is known about the role of the transcription factor nuclear factor-κB (NF-κB) in this process. We investigated the activities of the RELB proto-oncogene NF-κB subunit in human cholangiocytes and in mouse models of liver disease characterized by a ductular reaction., Methods: We obtained liver tissue samples from patients with primary sclerosing cholangitis, primary biliary cholangitis, hepatitis B or C virus infection, autoimmune hepatitis, alcoholic liver disease, or without these diseases (controls) from a tissue bank in Germany. Tissues were analyzed by immunohistochemistry for levels of RELB and lymphotoxin β (LTB). We studied mice with liver parenchymal cell (LPC)-specific disruption of the cylindromatosis (CYLD) lysine 63 deubiquitinase gene (Cyld), with or without disruption of Relb (Cyld ΔLPC mice and Cyld/Relb ΔLPC mice) and compared them with C57BL/6 mice (controls). Mice were fed 5-diethoxycarbonyl-1,4-dihydrocollidine (DDC) or standard chow diets to induce biliary injury or were given injections of CCl 4 to induce non-cholestatic liver fibrosis. Liver tissues were analyzed by histology, immunohistochemistry, immunoblots, in situ hybridization, and quantitative real-time polymerase chain reaction. Cholangiocytes were isolated from normal human liver, incubated with LTB receptor agonist, and transfected with small interfering RNAs to knock down RELB., Results: In liver tissues from patients with primary sclerosing cholangitis, primary biliary cholangitis, chronic infection with hepatitis B or C virus, autoimmune hepatitis, or alcoholic liver disease, we detected increased nuclear translocation of RELB and increased levels of LTB in cholangiocytes that formed reactive bile ducts compared with control liver tissues. Human cholangiocytes, but not those with RELB knockdown, proliferated with exposure to LTB. The phenotype of Cyld ΔLPC mice, which included ductular reaction, oval cell activation, and biliary fibrosis, was completely lost from Cyld/Relb ΔLPC mice. Compared with livers from control mice, livers from Cyld ΔLPC mice (but not Cyld/Relb ΔLPC mice) had increased levels of mRNAs encoding cytokines (LTB; CD40; and tumor necrosis factor superfamily [TNFSF] members TNFSF11 [RANKL], TNFSF13B [BAFF], and TNFSF14 [LIGHT]) produced by reactive cholangiocytes. However, these strains of mice developed similar levels of liver fibrosis in response to CCl 4 exposure. Cyld ΔLPC mice and Cyld/Relb ΔLPC mice had improved liver function on the DDC diet compared with control mice fed the DDC diet., Conclusion: Reactive bile ducts in patients with chronic liver diseases have increased levels of LTB and nuclear translocation of RELB. RELB is required for the ductular reaction and development of biliary fibrosis in Cyld ΔLPC mice. Deletion of RELB and CYLD from LPCs protects mice from DDC-induced cholestatic liver fibrosis., (Copyright © 2019 AGA Institute. Published by Elsevier Inc. All rights reserved.)