Tumor-colonized Streptococcus mutans metabolically reprograms tumor microenvironment and promotes oral squamous cell carcinoma.

BACKGROUND: Oral squamous cell carcinoma (OSCC) remains a major death cause in head and neck cancers, but the exact pathogenesis mechanisms of OSCC are largely unclear. RESULTS: Saliva derived from OSCC patients but not healthy controls (HCs) significantly promotes OSCC development and progression in rat models, and metabolomic analyses reveal saliva of OSCC patients but not HCs and OSCC tissues but not adjacent non-tumor tissues contain higher levels of kynurenic acid (KYNA). Furthermore, large amounts of Streptococcus mutans (S. mutans) colonize in OSCC tumor tissues, and such intratumoral S. mutans mediates KYNA overproductions via utilizing its protein antigen c (PAc). KYNA shifts the cellular types in the tumor microenvironment (TME) of OSCC and predominantly expedites the expansions of S100a8highS100a9high neutrophils to produce more interleukin 1ß (IL-1ß), which further expands neutrophils and induces CD8 + T cell exhaustion in TME and therefore promotes OSCC. Also, KYNA compromises the therapeutic effects of programmed cell death ligand 1 (PD-L1) and IL-1ß blockades in oral carcinogenesis model. Moreover, KYNA-mediated immunosuppressive program and aryl hydrocarbon receptor (AHR) expression correlate with impaired anti-tumor immunity and poorer survival of OSCC patients. CONCLUSIONS: Thus, aberration of oral microbiota and intratumoral colonization of specific oral bacterium such as S. mutans may increase the production of onco-metabolites, exacerbate the oral mucosal carcinogenesis, reprogram a highly immunosuppressive TME, and promote OSCC, highlighting the potential of interfering with oral microbiota and microbial metabolism for OSCC preventions and therapeutics. Video Abstract.

Single-Cell RNA-Sequencing of Soybean Reveals Transcriptional Changes and Antiviral Functions of GmGSTU23 and GmGSTU24 in Response to Soybean Mosaic Virus.

Soybean mosaic virus (SMV) stands as a prominent and widespread threat to soybean (Glycine max L. Merr.), the foremost legume crop globally. Attaining a thorough comprehension of the alterations in the transcriptional network of soybeans in response to SMV infection is imperative for a profound insight into the mechanisms of viral pathogenicity and host resistance. In this investigation, we isolated 50 294 protoplasts from the newly developed leaves of soybean plants subjected to both SMV infection and mock inoculation. Subsequently, we utilized single-cell RNA sequencing (scRNA-seq) to construct the transcriptional landscape at a single-cell resolution. Nineteen distinct cell clusters were identified based on the transcriptomic profiles of scRNA-seq. The annotation of three cell types-epidermal cells, mesophyll cells, and vascular cells-was established based on the expression of orthologs to reported marker genes in Arabidopsis thaliana. The differentially expressed genes between the SMV- and mock-inoculated samples were analyzed for different cell types. Our investigation delved deeper into the tau class of glutathione S-transferases (GSTUs), known for their significant contributions to plant responses against abiotic and biotic stress. A total of 57 GSTU genes were identified by a thorough genome-wide investigation in the soybean genome G. max Wm82.a4.v1. Two specific candidates, GmGSTU23 and GmGSTU24, exhibited distinct upregulation in all three cell types in response to SMV infection, prompting their selection for further research. The transient overexpression of GmGSTU23 or GmGSTU24 in Nicotiana benthamiana resulted in the inhibition of SMV infection, indicating the antiviral function of soybean GSTU proteins.

Fully automated hybrid approach on conventional MRI for triaging clinically significant liver fibrosis: A multi-center cohort study.

Establishing reliable noninvasive tools to precisely diagnose clinically significant liver fibrosis (SF, ≥F2) remains an unmet need. We aimed to build a combined radiomics-clinic (CoRC) model for triaging SF and explore the additive value of the CoRC model to transient elastography-based liver stiffness measurement (FibroScan, TE-LSM). This retrospective study recruited 595 patients with biopsy-proven liver fibrosis at two centers between January 2015 and December 2021. At Center 1, the patients before December 2018 were randomly split into training (276) and internal test (118) sets, the remaining were time-independent as a temporal test set (96). Another data set (105) from Center 2 was collected for external testing. Radiomics scores were built with selected features from Deep learning-based (ResUNet) automated whole liver segmentations on MRI (T2FS and delayed enhanced-T1WI). The CoRC model incorporated radiomics scores and relevant clinical variables with logistic regression, comparing routine approaches. Diagnostic performance was evaluated by the area under the receiver operating characteristic curve (AUC). The additive value of the CoRC model to TE-LSM was investigated, considering necroinflammation. The CoRC model achieved AUCs of 0.79 (0.70, 0.86), 0.82 (0.73, 0.89), and 0.81 (0.72-0.91), outperformed FIB-4, APRI (all p < 0.05) in the internal, temporal, and external test sets and maintained the discriminatory power in G0-1 subgroups (AUCs range, 0.85-0.86; all p < 0.05). The AUCs of joint CoRC-LSM model were 0.86 (0.79-0.94), and 0.81 (0.72-0.90) in the internal and temporal sets (p = 0.01). The CoRC model was useful for triaging SF, and may add value to TE-LSM.

Electrical stimulation induced pre-vascularization of engineered dental pulp tissue.

Vascularization is a key step to achieve pulp tissue regeneration and in vitro pre-vascularized dental pulp tissue could be applied as a graft substitute for dental pulp tissue repair. In this study, human dental pulp stem cells (DPSCs) and human umbilical vein endothelial cells (hUVECs) were co-cultured in 3D Matrigel and 150 mV/mm electric fields (EFs) were used to promote the construction of pre-vascularized dental pulp tissue. After optimizing co-cultured ratio of two cell types, immunofluorescence staining, and live/dead detection were used to investigate the effect of EFs on cell survival, differentiation and vessel formation in 3D engineered dental pulp tissue. RNA sequencing was used to investigate the potential molecular mechanisms by which EF regulates vessel formation in 3D engineered dental pulp tissue. Here we identified that EF-induced pre-vascularized engineered dental pulp tissue not only had odontoblasts, but also had a rich vascular network, and smooth muscle-like cells appeared around the blood vessels. The GO enrichment analysis showed that these genes were significantly enriched in regulation of angiogenesis, cell migration and motility. The most significant term of the KEGG pathway analysis were NOTCH signaling pathway and Calcium signaling pathway etc. The PPI network revealed that NOTCH1 and IL-6 were central hub genes. Our study indicated that EFs significantly promoted the maturation and stable of blood vessel in 3D engineered pulp tissue and provided an experimental basis for the application of EF in dental pulp angiogenesis and regeneration.

Exploring Novel Fentanyl Analogues Using a Graph-Based Transformer Model.

The structures of fentanyl and its analogues are easy to be modified and few types have been included in database so far, which allow criminals to avoid the supervision of relevant departments. This paper introduces a molecular graph-based transformer model, which is combined with a data augmentation method based on substructure replacement to generate novel fentanyl analogues. 140,000 molecules were generated, and after a set of screening, 36,799 potential fentanyl analogues were finally obtained. We calculated the molecular properties of 36,799 potential fentanyl analogues. The results showed that the model could learn some properties of original fentanyl molecules. We compared the generated molecules from transformer model and data augmentation method based on substructure replacement with those generated by the other two molecular generation models based on deep learning, and found that the model in this paper can generate more novel potential fentanyl analogues. Finally, the findings of the paper indicate that transformer model based on molecular graph helps us explore the structure of potential fentanyl molecules as well as understand distribution of original molecules of fentanyl.

Micro-environment triple-responsive hyaluronic acid hydrogel dressings to promote antibacterial activity, collagen deposition, and angiogenesis for diabetic wound healing.

The clinical treatment of chronic diabetic wounds is a long-standing thorny issue. Strategies targeting the diabetic micro-environment have been developed to promote wound healing. However, it remains challenging to reverse the adverse conditions and re-activate tissue regeneration and angiogenesis. In this work, we develop injectable hydrogels that are responsive to acidic conditions, reactive oxygen species (ROS), and high glucose levels in a diabetic wound micro-environment to sustainably deliver tannic acid (TA) to augment antibacterial, anti-inflammatory, and anti-oxidative activities. This triple-responsive mechanism is designed by introducing dynamic acylhydrazone and phenylboronic ester bonds to crosslink modified hyaluronic acid (HA) chains. At a diabetic wound, the acylhydrazone bonds may be hydrolyzed at low pH. Meanwhile, glucose may compete with TA, and ROS may oxidize the C-B bond to release TA. Thus, sustained release of TA is triggered by the diabetic micro-environment. The released TA effectively scavenges ROS and kills bacteria. In vivo experiments on diabetic mice demonstrate that the hydrogel dressing highly promotes angiogenesis and extracellular matrix (ECM) deposition, leading to eventual full healing of diabetic skin wounds. This micro-environment-triggered triple-responsive drug release provides a promising method for chronic diabetic wound healing.

Tandem Vinylogous Aldol and Intramolecular [2 + 2] Cycloaddition toward Benzocyclobutenes by UV Light Photocatalysis.

A facile and efficient radical tandem vinylogous aldol and intramolecular [2 + 2] cycloaddition reaction for direct synthesis of cyclobutane-containing benzocyclobutenes (BCBs) under extremely mild conditions without using any photocatalysts is reported. This approach exhibited definite compatibility with functional groups and afforded new BCBs with excellent regioselectivity and high yields. Moreover, detailed mechanism studies were carried out both experimentally and theoretically. The readily accessible, low-cost, and ecofriendly nature of the developed strategy will endow it with attractive applications in organic and medicinal chemistry.

A Comparative Study of Removal of Acid Red 27 by Adsorption on Four Different Chitosan Morphologies.

To investigate the relationship between structures and adsorption properties, four different morphologies of chitosan, with hydrogel (CSH), aerogel (CSA), powder (CSP), and electrospinning nanofiber (CSEN) characteristics, were employed as adsorbents for the removal of Acid Red 27. The structures and morphologies of the four chitosan adsorbents were characterized with SEM, XRD, ATR-FTIR, and BET methods. The adsorption behaviors and mechanisms of the four chitosan adsorbents were comparatively studied. All adsorption behaviors exhibited a good fit with the pseudo-second-order kinetic model (R2 > 0.99) and Langmuir isotherm model (R2 > 0.99). Comparing the adsorption rates and the maximum adsorption capacities, the order was CSH > CSA > CSP > CSEN. The maximum adsorption capacities of CSH, CSA, CSP, and CSEN were 2732.2 (4.523), 676.7 (1.119), 534.8 (0.885), and 215.5 (0.357) mg/g (mmol/g) at 20 °C, respectively. The crystallinities of CSH, CSA, CSP, and CSEN were calculated as 0.41%, 6.97%, 8.76%, and 39.77%, respectively. The crystallinity of the four chitosan adsorbents was the main factor impacting the adsorption rates and adsorption capacities, compared with the specific surface area. With the decrease in crystallinity, the adsorption rates and capacities of the four chitosan adsorbents increased gradually under the same experimental conditions. CSH with a low crystallinity and large specific surface area resulted in the highest adsorption rate and capacity.

Peripheral Mechanisms of Mechanical Itch.

Mechanical itch, which is defined as an itch sensation caused by innocuous mechanical force, may warn of the potential risk in the skin. The increased mechanosensitivity in sensory neurons may cause scratch-induced itch and promote the transition from acute itch to chronic itch. Recent studies have not only expanded our knowledge about the neuronal circuits in the CNS but have also highlighted the importance of the peripheral epithelia-immune-neuronal crosstalk in the development of mechanical itch. In this review, we will summarize related findings about the molecular and cellular mechanisms of mechanical itch in the skin.

bHLH transcription factors Hes1, Ascl1 and Oligo2 exhibit different expression patterns in the process of physiological electric fields-induced neuronal differentiation.

BACKGROUND: Recent studies have shown that the expression of bHLH transcription factors Hes1, Ascl1, and Oligo2 has an oscillating balance in neural stem cells (NSCs) to maintain their self-proliferation and multi-directional differentiation potential. This balance can be disrupted by exogenous stimulation. Our previous work has identified that electrical stimulation could induce neuronal differentiation of mouse NSCs. METHODS: To further evaluate if physiological electric fields (EFs)-induced neuronal differentiation is related to the expression patterns of bHLH transcription factors Hes1, Ascl1, and Oligo2, mouse embryonic brain NSCs were used to investigate the expression changes of Ascl1, Hes1 and Oligo2 in mRNA and protein levels during EF-induced neuronal differentiation. RESULTS: Our results showed that NSCs expressed high level of Hes1, while expression of Ascl1 and Oligo2 stayed at very low levels. When NSCs exited proliferation, the expression of Hes1 in differentiated cells began to decrease and oscillated at the low expression level. Oligo2 showed irregular changes in low expression level. EF-stimulation significantly increased the expression of Ascl1 at mRNA and protein levels accompanied by an increased percentage of neuronal differentiation. What's more, over-expression of Hes1 inhibited the neuronal differentiation induced by EFs. CONCLUSION: EF-stimulation directed neuronal differentiation of NSCs by promoting the continuous accumulation of Ascl1 expression and decreasing the expression of Hes1.

Canine distemper virus (CDV)-neutralizing activities of an anti-CDV canine-derived single-chain variable antibody fragment 4-15 (scFv 4-15) screened by phage display technology.

Canine distemper virus (CDV) is a highly contagious pathogen that causes severe diarrhea, fever and vomiting in domestic dogs, posing a serious threat to the dog breeding industry. Currently, there are no effective therapeutic agents for emergency treatment despite the availability of vaccines against CDV infection. Single-chain fragment variable (scFv) antibody has been demonstrated to effectively inhibit virus infections, suggesting a potential candidate as a therapeutic agent for canine distemper. In this study, a phage-displayed scFv library was constructed from the peripheral blood lymphocytes of dog immunized intramuscularly with live-attenuated CDV vaccine, and was subjected to four rounds of pannings against CDV. Subsequent indirect enzyme-linked immunosorbent assay screening revealed high-affinity scFv antibodies specific to CDV, and indirect immunofluorescence assay screening revealed CDV-neutralizing activity of scFv antibodies. Our results showed that a scFv antibody 4-15 (scFv 4-15) with high-affinity binding to CDV and neutralizing activity against CDV was obtained, which displayed effective therapeutic potential in vivo for dogs challenged with a lethal dose of CDV. Conclusively, the scFv 4-15 with high-affinity binding and neutralizing activity to CDV that was obtained by phage display technology provides a promising candidate for the therapeutic agents against CDV infection.

Identification and validation of m6A RNA regulatory network in pulpitis.

BACKGROUND: N6-methyladenosine (m6A) RNA modification regulators play an important role in many human diseases, and its abnormal expression can lead to the occurrence and development of diseases. However, their significance in pulpitis remains largely unknown. Here, we sought to identify and validate the m6A RNA regulatory network in pulpitis. METHODS: Gene expression data for m6A regulators in human pulpitis and normal pulp tissues from public GEO databases were analyzed. Bioinformatics analysis including Gene ontology (GO) functional, and Kyoto encyclopedia of genes and genomes (KEGG) pathway analyses were performed by R package, and Cytoscape software was used to study the role of m6A miRNA-mRNA regulatory network in pulpitis. Quantitative real-time PCR (qRT-PCR) was performed to validate the expression of key m6A regulators in collected human pulpitis specimens. RESULTS: Differential genes between pulpitis and normal groups were found from the GEO database, and further analysis found that there were significant differences in the m6A modification-related genes ALKBH5, METTL14, METTL3, METTL16, RBM15B and YTHDF1. And their interaction relationships and hub genes were determined. The hub m6A regulator targets were enriched in immune cells differentiation, glutamatergic synapse, ephrin receptor binding and osteoclast differentiation in pulpitis. Validation by qRT-PCR showed that the expression of methylases METTL14 and METTL3 was decreased, thus these two genes may play a key role in pulpitis. CONCLUSION: Our study identified and validated the m6A RNA regulatory network in pulpitis. These findings will provide valuable resource to guide the mechanistic and therapeutic analysis of the role of key m6A modulators in pulpitis.

Visible-Light-Mediated Catalyst-Free [2+2] Cycloaddition Reaction for Dihydrocyclobuta[b]naphthalene-3,8-diones Synthesis under Mild Conditions.

A facile and efficient visible-light-mediated method for directly converting 1,4-naphthoquinones into dihydrocyclo-buta[b]naphthalene-3,8-diones (DHCBNDOs) under mild and clean conditions without using any photocatalysts is reported. This approach exhibited favorable compatibility with functional groups and afforded a series of DHCBNDOs with excellent regioselectivity and high yields. Moreover, detailed mechanism studies were carried out both experimentally and theoretically. The readily accessible, low-cost and ecofriendly nature of the developed strategy will endow it with attractive applications in organic and medicinal chemistry.

Identification and Genome Characterization of a Dahlia Common Mosaic Virus Isolate from China.

Dahlia (Dahlia variabilis) is a widely cultivated ornamental and medicinal plant in China. Recently, dahlia plants with symptoms of leaf mottling and distortion were collected in Hohhot, Inner Mongolia, China. The presence of dahlia common mosaic virus (DCMV), an unassigned species in the genus Caulimovirus, was confirmed by high-throughput sequencing. Three fragments of DCMV Inner Mongolia isolate (DCMV-IN) were PCR-amplified with specific primers, sequenced and assembled into the complete genome sequence with a GenBank accession number of OR494328. The double-stranded circular DNA genome of DCMV-IN consists of 7949 bp and contains six open reading frames (ORFs). Sequence analysis showed that DCMV-IN shared high sequence identities with other DCMV isolates available in the GenBank database. Phylogenetic analysis of DCMV isolates and other representative caulimoviruses based on genome sequence clustered four DCMV isolates to a single branch which was closest to dahlia mosaic virus (DMV). No recombination event was detected among the four DCMV isolates.

Up-regulated lncRNA CYLD as a ceRNA of miR-2383 facilitates bovine viral diarrhea virus replication by promoting CYLD expression to counteract RIG-I-mediated type-I IFN production.

Bovine viral diarrhea virus (BVDV) is one of the most important pathogens of cattle, causing numerous economic losses to the cattle industry. To date, many potential mechanisms of BVDV evading or subverting innate immunity are still unknown. In this study, an lnc-CYLD/miR-2383/CYLD axis involved in BVDV-host interactions was screened from RNA-seq-based co-expression networks analysis of long noncoding RNAs, microRNAs and mRNAs in BVDV-infected bovine cells, and underlying mechanisms of lnc-CYLD/miR-2383/CYLD axis regulating BVDV replication were explored. Results showed that BVDV-induced up-regulation of the lnc-CYLD competed for binding to the miR-2383, and then promoted CYLD expression, thereby inhibiting RIG-I-mediated type-I interferon (IFN) production, which was subsequently confirmed by treatment with lnc-CYLD overexpression and miR-2383 inhibitor. However, miR-2383 transfection and small interfering RNA-mediated lnc-CYLD knockdown inhibited CYLD expression and enhanced RIG-I-mediated type-I IFN production, inhibiting BVDV replication. In addition, interaction relationship between lnc-CYLD and miR-2383, and colocalization relationship of lnc-CYLD, miR-2383 and CYLD were confirmed by dual-luciferase assay and in situ hybridization assay. Conclusively, up-regulation of the lnc-CYLD as a competing endogenous RNA binds to the miR-2383 to reduce inhibitory effect of the miR-2383 on the CYLD expression, playing an important role in counteracting type-I IFN-dependent antiviral immunity to facilitate BVDV replication.

Targeting CD36 determines nicotine derivative NNK-induced lung adenocarcinoma carcinogenesis.

Smoking carcinogen nicotine-derived nitrosamine ketone (NNK) is the most potent contributor to lung adenocarcinoma (LUAD) development, but the mechanism has not been fully elucidated. Here, we reported that fatty acid translocase CD36 was significantly overexpressed in both human LUAD tissues and NNK-induced A/J mice LUAD tumors. The overexpressed CD36 was positively correlated with Src kinase activation, smoking status, metastasis, and worse overall survival of patients with smoking history. Upon NNK binding with α7 nicotinic acetylcholine receptor (α7nAChR), sarcolemmal CD36 was increased and it interacted with surface α7nAChR and cytosol Src simultaneously, which in turn activated Src and downstream pro-carcinogenic kinase ERK1/2 and Akt, and finally caused LUAD cells to form subcutaneous and pulmonary metastatic tumors. This process could be blocked by CD36 knockdown and CD36 irreversible inhibitor SSO. Furthermore, the effect of NNK was inhibited obviously in CD36-/- A/J mice. Thus, targeting CD36 may provide a breakthrough therapy of LUAD.

The activated CD36-Src axis promotes lung adenocarcinoma cell proliferation and actin remodeling-involved metastasis in high-fat environment.

Obesity/overweight and lipid metabolism disorders have become increased risk factors for lung cancer. Fatty acid translocase CD36 promotes cellular uptake of fatty acids. Whether and how CD36 facilitates lung adenocarcinoma (LUAD) growth in high-fat environment is unknown. Here, we demonstrated that palmitic acid (PA) or high-fat diet (HFD) promoted LUAD cell proliferation and metastasis in a CD36-dependent manner. Mechanistically, CD36 translocated from cytoplasm to cell membrane and interacted with Src kinase upon PA stimulation in human LUAD cells. Akt and ERK, downstream of Src, were then activated to mediate LUAD cell proliferation and metastasis. Furthermore, PA treatment promoted CD36 sarcolemmal translocation, where it activated Rac1 and upregulated MMP-9 through Src-Akt/ERK pathway, resulting in redistribution of cortactin, N-WASP and Arp2/3, and finally led to occurrence of finger-like protrusions of actin on cell surface to enhance cell metastasis. Compared with normal-chew diet (NCD) mice, the HFD group exhibited higher level of blood free fatty acid (FFA) and cholesterol (TC), developed larger xenograft LUAD tumors and enhanced tumor cell metastatic potential, which were accompanied by obvious sarcolemmal actin remodeling and were blocked by simultaneous CD36 knockdown in LUAD cells. Consistently, xenografted and tail vein-injected scramble-RNA-A549 cells but not CD36-shRNA-A549 in HFD mice formed metastatic LUAD tumors on the lung. CD36 inhibitor SSO significantly inhibited LUAD cell metastasis to the lung. Collectively, CD36 initiates Src signaling to promote LUAD cell proliferation and actin remodeling-involved metastasis under high-fat environment. Our study provides the new insights that CD36 is a valid target for LUAD therapy.

Near-Infrared Fluorescent Probe for the In Situ Visualization of Oxidative Stress in the Brains of Neuroinflammatory and Schizophrenic Mice.

Schizophrenia is a common mental disorder with unclear mechanisms. Oxidative stress and neuroinflammation play important roles in the pathological process of schizophrenia. Superoxide anion (O2•-) is an important oxidative stress biomarker in vivo. However, due to the existence of the blood-brain barrier (BBB), few near-infrared (NIR) fluorescent probes have been used for the sensing and detection of O2•- in the brain. With this research, we developed the first near-infrared fluorescent probe (named CT-CF3) for noninvasive detection of endogenous O2•- in the brain of mice. Enabling fluorescence monitoring of the dynamic changes in O2•- flux due to the prolonged activation of microglia in neuroinflamed and schizophrenic (SZ) mice brains, thereby providing direct evidence for the relationship between oxidative stress, neuroinflammation, and schizophrenia. Furthermore, we confirmed the O2•- burst in the brains of first-episode schizophrenic mice and assessed the effect of two atypical antipsychotic drugs (risperidone and olanzapine) on redox homeostasis.

Electrical stimulation induced structural 3D human engineered neural tissue with well-developed neuronal network and functional connectivity.

Objective.Three-dimensional (3D) neural tissue engineering is expected to provide new stride in developing neural disease models and functional substitutes to aid in the treatment of central nervous system injury. We have previously detailed an electrical stimulation (ES) system to generate 3D mouse engineered neural tissue (mENT)in vitro. However, ES-induced human ENT (hENT) has not previously been either investigated or identified in structural and functional manner. Here, we applied ES as a stimulator to regulate human neural stem cells in 3D Matrigel, explored the components and functional properties of hENTs.Approach.By immunofluorescence chemical staining and electron microscope imaging, we evaluated the effects of ES on (1) neuronal differentiation and maturation, (2) neurites outgrowth and alignment in hENT, (3) formation of synapses and myelin sheaths in hENT. We further investigated the formation of synaptic connections betweenex-vivo-fused mouse and human tissue. We used calcium imaging to detect activities of neurons in hENT culture.Results.ES could induce neuronal differentiation, the orderly growth of neurites and the maturation of neuron subtypes to construct a well-developed neuronal network with synapses and myelin sheaths. Most importantly, we discovered that raising extracellular K+concentration resulted the increasing neuronal excitability in the hENT, indicating electrical activities in neuronal cells.Significance.We applied ES to generate the organised 3D hENTs and identified them in both structural and functional manner.

An imaging-based machine learning model outperforms clinical risk scores for prognosis of cirrhotic variceal bleeding.

OBJECTIVES: To develop and validate a machine learning model based on contrast-enhanced CT to predict the risk of occurrence of the composite clinical endpoint (hospital-based intervention or death) in cirrhotic patients with acute variceal bleeding (AVB). METHODS: This retrospective study enrolled 330 cirrhotic patients with AVB between January 2017 and December 2020 from three clinical centers. Contrast-enhanced CT and clinical data were collected. Centers A and B were divided 7:3 into a training set and an internal test set, and center C served as a separate external test set. A well-trained deep learning model was applied to segment the liver and spleen. Then, we extracted 106 original features of the liver and spleen separately based on the Image Biomarker Standardization Initiative (IBSI). We constructed the Liver-Spleen (LS) model based on the selected radiomics features. The performance of LS model was evaluated by receiver operating characteristics and calibration curves. The clinical utility of models was analyzed using decision curve analyses (DCA). RESULTS: The LS model demonstrated the best diagnostic performance in predicting the composite clinical endpoint of AVB in patients with cirrhosis, with an AUC of 0.782 (95% CI 0.650-0.882) and 0.789 (95% CI 0.674-0.878) in the internal test and external test groups, respectively. Calibration curves and DCA indicated the LS model had better performance than traditional clinical scores. CONCLUSION: A novel machine learning model outperforms previously known clinical risk scores in assessing the prognosis of cirrhotic patients with AVB CLINICAL RELEVANCE STATEMENT: The Liver-Spleen model based on contrast-enhanced CT has proven to be a promising tool to predict the prognosis of cirrhotic patients with acute variceal bleeding, which can facilitate decision-making and personalized therapy in clinical practice. KEY POINTS: • The Liver-Spleen machine learning model (LS model) showed good performance in assessing the clinical composite endpoint of cirrhotic patients with AVB (AUC ≥ 0.782, sensitivity ≥ 80%). • The LS model outperformed the clinical scores (AUC ≤ 0.730, sensitivity ≤ 70%) in both internal and external test cohorts.