Treatment of Hemodialysis Vascular Access repeated Stenosis with Paclitaxel-Coated Balloon Angioplasty: A Retrospective Study.

Introduction To report our experience of angioplasty with Paclitaxel-coated balloon (PCB) versus common balloon (CB) for the treatment of repeated failing vascular access. Methods Retrospective, single-center analysis consisting of 88 patients treated with percutaneous transluminal angioplasty (PTA) in the period from October 2020 through December 2021. Patients were divided into two groups according to the type of treatment as Paclitaxel-coated balloon (n=41) and common balloon (n=47). We analyzed target lesion primary patency and vascular access primary patency for 6 months, and the rate of complications. Results There was no significant difference in the target lesion primary patency which was similar for 6 months between the two groups (PCB group vs. CB group at 1,3,6 months; 95.12% vs. 89.36%(P=0.55), 75.61% vs. 74.47%(P=0.90), 53.66% vs. 63.83%(P=0.33), respectively). Similarly, vascular access primary patency in the PCB group and CB group was 90.24% and 89.36%(P=0.83), respectively, at 1 month, 65.85% and 68.09%(P=0.82), respectively, at 3 months, 39.02% and 53.19%(P=0.18), respectively, at 6 months. There were no major complications after endovascular treatment. Conclusion Compared to common balloon angioplasty, paclitaxel-coated balloon angioplasty have no short-term patency benefit in the treatment of vascular access repeated stenosis.

Potential of PAQosome as a therapeutic target for hepatic fibrosis.

BACKGROUND AND AIM: The condition of hepatic fibrosis is hazardous. Therefore, it is vital that we investigate the mechanism of hepatic fibrosis to provide new targets for treatment. METHODS: Preliminary screening and research was carried out based on our prior results and our speculated role of the particle with quaternary structure arrangement (PAQosome) in hepatic fibrosis. The experiments were conducted using LX-2 or HepG2 cell lines by western blotting, quantitative real-time polymerase chain reaction, luciferase assays, and co-immunoprecipitation and were further validated in the Gene Expression Omnibus (GEO) database. RESULTS: We screened and proved that several subunits of the PAQosome regulate the development of liver fibrosis, including the asparagine synthetase domain-containing 1 upstream open reading frame (ASDURF), prefoldin subunit 4 (PFDN4), prefoldin subunit 5 (PFDN5), unconventional prefoldin RNA polymerase II subunit 5 interactor (URI1), and ubiquitously expressed prefoldin-like chaperone (UXT). ASDURF promotes hepatic fibrosis through the transforming growth factor-ß1 (TGFß1)/Sekelsky mothers against decapentaplegic homologue 3 (Smad3) and NF-κB signaling pathways. ASDURF regulates the expression of asparagine synthetase domain-containing 1 (ASNSD1). PFDN4, PFDN5, URI1, and UXT regulate cell proliferation through the PI3K/AKT pathway, and thus regulate liver fibrosis. A hepatic fibrosis score ≥ F2 was selected as the diagnostic criteria for hepatic fibrosis in the GSE96971 database. The area under the receiver operating characteristic curve of PFDN4, PFDN5, UXT, and ASNSD1 were 0.862 (confidence interval [CI]: 0.6588-1.000), 0.538 (CI: 0.224-0.853), 0.708 (CI: 0.449-0.966), and 0.831 (CI: 0.638-1.000), respectively. CONCLUSIONS: These findings demonstrate that the PAQosome is a brand new target for hepatic fibrosis therapy.

Recompensation in cirrhosis: unravelling the evolving natural history of nonalcoholic fatty liver disease.

Recompensation has gained increasing attention in the field of cirrhosis, particularly in chronic liver disease with a definite aetiology. The current global prevalence of obesity and nonalcoholic fatty liver disease (NAFLD) is increasing, but there is currently a lack of a clear definition for recompensation in NAFLD-related cirrhosis. Here, we provide an up-to-date perspective on the natural history of NAFLD, emphasizing the reversible nature of the disease, summarizing possible mechanisms underlying recompensation in NAFLD, discussing challenges that need to be addressed and outlining future research directions in the field. Recompensation is a promising goal in patients with NAFLD-related cirrhosis, and further studies are needed to explore its underlying mechanisms and uncover its clinical features.

Automated diagnosis of anterior cruciate ligament via a weighted multi-view network.

Objective: To build a three-dimensional (3D) deep learning-based computer-aided diagnosis (CAD) system and investigate its applicability for automatic detection of anterior cruciate ligament (ACL) of the knee joint in magnetic resonance imaging (MRI). Methods: In this study, we develop a 3D weighted multi-view convolutional neural network by fusing different views of MRI to detect ACL. The network is evaluated on two MRI datasets, the in-house MRI-ACL dataset and the publicly available MRNet-v1.0 dataset. In the MRI-ACL dataset, the retrospective study collects 100 cases, and four views per patient are included. There are 50 ACL patients and 50 normal patients, respectively. The MRNet-v1.0 dataset contains 1,250 cases with three views, of which 208 are ACL patients, and the rest are normal or other abnormal patients. Results: The area under the receiver operating characteristic curve (AUC) of the ACL diagnosis system is 97.00% and 92.86% at the optimal threshold for the MRI-ACL dataset and the MRNet-v1.0 dataset, respectively, indicating a high overall diagnostic accuracy. In comparison, the best AUC of the single-view diagnosis methods are 96.00% (MRI-ACL dataset) and 91.78% (MRNet-v1.0 dataset), and our method improves by about 1.00% and 1.08%. Furthermore, our method also improves by about 1.00% (MRI-ACL dataset) and 0.28% (MRNet-v1.0 dataset) compared with the multi-view network (i.e., MRNet). Conclusion: The presented 3D weighted multi-view network achieves superior AUC in diagnosing ACL, not only in the in-house MRI-ACL dataset but also in the publicly available MRNet-v1.0 dataset, which demonstrates its clinical applicability for the automatic detection of ACL.

Suppressive stroma-immune prognostic signature impedes immunotherapy in ovarian cancer and can be reversed by PDGFRB inhibitors.

BACKGROUND: As the most lethal gynecologic cancer, ovarian cancer (OV) holds the potential of being immunotherapy-responsive. However, only modest therapeutic effects have been achieved by immunotherapies such as immune checkpoint blockade. This study aims to propose a generalized stroma-immune prognostic signature (SIPS) to identify OV patients who may benefit from immunotherapy. METHODS: The 2097 OV patients included in the study were significant with high-grade serous ovarian cancer in the III/IV stage. The 470 immune-related signatures were collected and analyzed by the Cox regression and Lasso algorithm to generalize a credible SIPS. Correlations between the SIPS signature and tumor microenvironment were further analyzed. The critical immunosuppressive role of stroma indicated by the SIPS was further validated by targeting the major suppressive stroma component (CAFs, Cancer-associated fibroblasts) in vitro and in vivo. With four machine-learning methods predicting tumor immune subtypes, the stroma-immune signature was upgraded to a 23-gene signature. RESULTS: The SIPS effectively discriminated the high-risk individuals in the training and validating cohorts, where the high SIPS succeeded in predicting worse survival in several immunotherapy cohorts. The SIPS signature was positively correlated with stroma components, especially CAFs and immunosuppressive cells in the tumor microenvironment, indicating the critical suppressive stroma-immune network. The combination of CAFs' marker PDGFRB inhibitors and frontline PARP inhibitors substantially inhibited tumor growth and promoted the survival of OV-bearing mice. The stroma-immune signature was upgraded to a 23-gene signature to improve clinical utility. Several drug types that suppress stroma-immune signatures, such as EGFR inhibitors, could be candidates for potential immunotherapeutic combinations in ovarian cancer. CONCLUSIONS: The stroma-immune signature could efficiently predict the immunotherapeutic sensitivity of OV patients. Immunotherapy and auxiliary drugs targeting stroma could enhance immunotherapeutic efficacy in ovarian cancer.

Loss of HRD functional phenotype impedes immunotherapy and can be reversed by HDAC inhibitor in ovarian cancer.

In recent years, homologous recombination deficiency (HRD) has not achieved the expected substantial promotion of immunotherapeutic efficacy in ovarian cancer. This study aims to explore the role of HRD functional phenotype as a powerful biomarker in identifying HRD patients who may benefit from immunotherapy. HRD functional phenotype, namely HRD-EXCUTE, was defined as the average level of the 15 hub genes upregulated in HRD ovarian cancer. A decision tree was plotted to evaluate the critical role of HRD-EXCUTE in HRD patients. Agents inducing HRD-EXCUTE were identified by CMAP web (Connectivity Map). The mechanisms and immunotherapeutic effect of PARPi and HDACi in promoting HRD-EXCUTE was examined in vitro and in vivo. The decision tree plotted on the basis of HRD and HRD-EXCUTE indicated the HRD patients without the HRD functional phenotype were largely unresponsive to immunotherapy, which was validated by the immunotherapeutic cohorts. Furthermore, loss of HRD-EXCUTE in the HRD patients attenuated immunogenicity and inhibited immune cells in tumor microenvironment. Moreover, Niraparib combined with Entinostat induced HRD-EXCUTE by activating the cGAS-STING pathway and increasing the histone acetylation. The combination therapy could enhance the cytotoxicity of immune cells, and promote pro-immune cells infiltrating into ascites, resulting in inhibited ovarian cancer growth. The HRD functional phenotype HRD-EXCUTE was set up as a potent biomarker to identify whether HRD patients can benefit from immunotherapy. Loss of HRD-EXCUTE in HRD patients were largely insensitive to immunotherapy. The combination of PARPi with HDACi could improve the efficacy of the PARPi-based immunotherapy in ovarian cancer by augmenting the HRD functional phenotype.

Tubastatin A potently inhibits GPX4 activity to potentiate cancer radiotherapy through boosting ferroptosis.

Ferroptosis, an iron-dependent lipid peroxidation-driven programmed cell death, is closely related to cancer therapy. The development of druggable ferroptosis inducers and their rational application in cancer therapy are critical. Here, we identified Tubastatin A, an HDAC6 inhibitor as a novel druggable ferroptosis inducer through large-scale drug screening. Tubastatin A directly bonded to GPX4 and inhibited GPX4 enzymatic activity through biotin-linked Tubastatin A putdown and LC/MS analysis, which is independent of its inhibition of HDAC6. In addition, our results showed that radiotherapy not only activated Nrf2-mediated GPX4 transcription but also inhibited lysosome-mediated GPX4 degradation, subsequently inducing ferroptosis tolerance and radioresistance in cancer cells. Tubastatin A overcame ferroptosis resistance and radioresistance of cancer cells by inhibiting GPX4 enzymatic activity. More importantly, Tubastatin A has excellent bioavailability, as demonstrated by its ability to significantly promote radiotherapy-induced lipid peroxidation and tumour suppression in a mouse xenograft model. Our findings identify a novel druggable ferroptosis inducer, Tubastatin A, which enhances radiotherapy-mediated antitumor effects. This work provides a compelling rationale for the clinical evaluation of Tubastatin A, especially in combination with radiotherapy.

Novel urinary protein panels for the non-invasive diagnosis of non-alcoholic fatty liver disease and fibrosis stages.

BACKGROUND & AIMS: There is an unmet clinical need for non-invasive tests to diagnose non-alcoholic fatty liver disease (NAFLD) and individual fibrosis stages. We aimed to test whether urine protein panels could be used to identify NAFLD, NAFLD with fibrosis (stage F ≥ 1) and NAFLD with significant fibrosis (stage F ≥ 2). METHODS: We collected urine samples from 100 patients with biopsy-confirmed NAFLD and 40 healthy volunteers, and proteomics and bioinformatics analyses were performed in this derivation cohort. Diagnostic models were developed for detecting NAFLD (UPNAFLD model), NAFLD with fibrosis (UPfibrosis model), or NAFLD with significant fibrosis (UPsignificant fibrosis model). Subsequently, the derivation cohort was divided into training and testing sets to evaluate the efficacy of these diagnostic models. Finally, in a separate independent validation cohort of 100 patients with biopsy-confirmed NAFLD and 45 healthy controls, urinary enzyme-linked immunosorbent assay analyses were undertaken to validate the accuracy of these new diagnostic models. RESULTS: The UPfibrosis model and the UPsignificant fibrosis model showed an AUROC of .863 (95% CI: .725-1.000) and 0.858 (95% CI: .712-1.000) in the training set; and .837 (95% CI: .711-.963) and .916 (95% CI: .825-1.000) in the testing set respectively. The UPNAFLD model showed an excellent diagnostic performance and the area under the receiver operator characteristic curve (AUROC) exceeded .90 in the derivation cohort. In the independent validation cohort, the AUROC for all three of the above diagnostic models exceeded .80. CONCLUSIONS: Our newly developed models constructed from urine protein biomarkers have good accuracy for non-invasively diagnosing liver fibrosis in NAFLD.

Hepatocyte apoptosis fragment product cytokeratin-18 M30 level and non-alcoholic steatohepatitis risk diagnosis: an international registry study.

BACKGROUND: Liver biopsy for the diagnosis of non-alcoholic steatohepatitis (NASH) is limited by its inherent invasiveness and possible sampling errors. Some studies have shown that cytokeratin-18 (CK-18) concentrations may be useful in diagnosing NASH, but results across studies have been inconsistent. We aimed to identify the utility of CK-18 M30 concentrations as an alternative to liver biopsy for non-invasive identification of NASH. METHODS: Individual data were collected from 14 registry centers on patients with biopsy-proven non-alcoholic fatty liver disease (NAFLD), and in all patients, circulating CK-18 M30 levels were measured. Individuals with a NAFLD activity score (NAS) ≥5 with a score of ≥1 for each of steatosis, ballooning, and lobular inflammation were diagnosed as having definite NASH; individuals with a NAS ≤2 and no fibrosis were diagnosed as having non-alcoholic fatty liver (NAFL). RESULTS: A total of 2571 participants were screened, and 1008 (153 with NAFL and 855 with NASH) were finally enrolled. Median CK-18 M30 levels were higher in patients with NASH than in those with NAFL (mean difference 177 U/L; standardized mean difference [SMD]: 0.87 [0.69-1.04]). There was an interaction between CK-18 M30 levels and serum alanine aminotransferase, body mass index (BMI), and hypertension ( P  < 0.001, P  = 0.026 and P  = 0.049, respectively). CK-18 M30 levels were positively associated with histological NAS in most centers. The area under the receiver operating characteristics (AUROC) for NASH was 0.750 (95% confidence intervals: 0.714-0.787), and CK-18 M30 at Youden's index maximum was 275.7 U/L. Both sensitivity (55% [52%-59%]) and positive predictive value (59%) were not ideal. CONCLUSION: This large multicenter registry study shows that CK-18 M30 measurement in isolation is of limited value for non-invasively diagnosing NASH.

Advances in tumor regression patterns and safe distance of distal resection margin after neoadjuvant therapy for rectal cancer / 中华胃肠外科杂志

Neoadjuvant therapy has been widely applied in the treatment of rectal cancer, which can shrink tumor size, lower tumor staging and improve the prognosis. It has been the standard preoperative treatment for patients with locally advanced rectal cancer. The efficacy of neoadjuvant therapy for rectal cancer patients varies between individuals, and the results of tumor regression are obviously different. Some patients with good tumor regression even achieve pathological complete response (pCR). Tumor regression is of great significance for the selection of surgical regimes and the determination of distal resection margin. However, few studies focus on tumor regression patterns. Controversies on the safe distance of distal resection margin after neoadjuvant treatment still exist. Therefore, based on the current research progress, this review summarized the main tumor regression patterns after neoadjuvant therapy for rectal cancer, and classified them into three types: tumor shrinkage, tumor fragmentation, and mucin pool formation. And macroscopic regression and microscopic regression of tumors were compared to describe the phenomenon of non-synchronous regression. Then, the safety of non-surgical treatment for patients with clinical complete response (cCR) was analyzed to elaborate the necessity of surgical treatment. Finally, the review studied the safe surgical resection range to explore the safe distance of distal resection margin.

Indole alkaloids fusarindoles A-E from marine-derived fungus Fusarium equiseti LJ-1.

Five undescribed indole alkaloids, fusarindoles A-E, together with seven known compounds were obtained from the marine-derived fungus Fusarium equiseti LJ-1. Their chemical structures and absolute configurations were determined by comprehensive analysis of the NMR, HRMS, UV, IR, ECD calculation and single-crystal X-ray diffraction data. The possible biosynthetic pathways of fusarindoles C-E were proposed. The cytotoxicities of eleven compounds, including fusarindoles A-E and six known compounds, against five human cancer cell lines A549, CNE2, SUNE1, HepG2 and QGY7701 were evaluated.

Machine learning algorithms based on proteomic data mining accurately predicting the recurrence of hepatitis B-related hepatocellular carcinoma.

BACKGROUND AND AIM: Over 10% of hepatocellular carcinoma (HCC) cases recur each year, even after surgical resection. Currently, there is a lack of knowledge about the causes of recurrence and the effective prevention. Prediction of HCC recurrence requires diagnostic markers endowed with high sensitivity and specificity. This study aims to identify new key proteins for HCC recurrence and to build machine learning algorithms for predicting HCC recurrence. METHODS: The proteomics data for analysis in this study were obtained from the Clinical Proteomics Tumor Analysis Consortium (CPTAC) database. We analyzed different proteins based on cases with or without recurrence of HCC. Survival analysis, Cox regression analysis, and area under the ROC curves (AUROC > 0.7) were used to screen for more significant differential proteins. Predictive models for HCC recurrence were developed using four machine learning algorithms. RESULTS: A total of 690 differentially expressed proteins between 50 relapsed and 77 non-relapsed hepatitis B-related HCC patients were identified. Seven of these proteins had an AUROC > 0.7 for 5-year survival in HCC, including BAHCC1, ESF1, RAP1GAP, RUFY1, SCAMP3, STK3, and TMEM230. Among the machine learning algorithms, the random forest algorithm showed the highest AUROC values (AUROC: 0.991, 95% CI 0.962-0.999) for identifying HCC recurrence, followed by the support vector machine (AUROC: 0.893, 95% Cl 0.824-0.956), the logistic regression (AUROC: 0.774, 95% Cl 0.672-0.868), and the multi-layer perceptron algorithm (AUROC: 0.571, 95% Cl 0.459-0.682). CONCLUSIONS: Our study identifies seven novel proteins for predicting HCC recurrence and the random forest algorithm as the most suitable predictive model for HCC recurrence.

The impact of direct-acting antivirals on quality of life in patients with hepatitis C virus infection: a meta-analysis.

INTRODUCTION AND OBJECTIVES: It is well known that the quality of life (QoL) of patients with chronic hepatitis C (HCV) is lower than that of the general population and that therapy with direct-acting antivirals (DAA) for HCV is safe and effective. However, data on the QoL of patients are scanty. The purpose of this study was to assess the effect of DAA drugs on patients' QoL. METHODS: The literature included in this meta-analysis was due in March 2021. The random effect model of heterogeneous data and the fixed effect model of homogeneous data were used to analyze the data. QoL had to be evaluated using the Short Form Health Survey (SF-36) questionnaire with at least one measure at baseline (T0) and one measure at 12 weeks (T12) or 24 weeks (T24) after the end of therapy. The meta-analysis included eight studies, which involved 1,619 patients. RESULTS: At T12, the meta-analysis showed all items of the SF-36 questionnaire improved from the pretreatment to post-treatment period and reached statistical significance (p < 0.05) except for the bodily pain (mean difference: 1.16, 95%CI -0.43-2.74) and role limitations-emotional (mean difference: 4.10, 95%CI -1.32-9.52). However, after subgroup analysis (whether ribavirin was being used or not), the bodily pain domain (mean difference: 3.34, 95%CI 1.03-5.65) became statistically significant again. At T24, the results indicated that all items of the SF-36 questionnaire improved from the pretreatment to the post-treatment period and reached statistical significance (p < 0.05) except for the role limitations-emotional domain (mean difference: 4.50, 95%CI -2.66-11.66). CONCLUSIONS: There is evidence indicating that DAA therapy is accompanied by an improvement in QoL. Patients receiving DAA medication have a clinically relevant improvement in most domains of the SF-36 questionnaire at T12 or T24, except for a few aspects including role limitations-emotional.

PKCßII phosphorylates ACSL4 to amplify lipid peroxidation to induce ferroptosis.

The accumulation of lipid peroxides is recognized as a determinant of the occurrence of ferroptosis. However, the sensors and amplifying process of lipid peroxidation linked to ferroptosis remain obscure. Here we identify PKCßII as a critical contributor of ferroptosis through independent genome-wide CRISPR-Cas9 and kinase inhibitor library screening. Our results show that PKCßII senses the initial lipid peroxides and amplifies lipid peroxidation linked to ferroptosis through phosphorylation and activation of ACSL4. Lipidomics analysis shows that activated ACSL4 catalyses polyunsaturated fatty acid-containing lipid biosynthesis and promotes the accumulation of lipid peroxidation products, leading to ferroptosis. Attenuation of the PKCßII-ACSL4 pathway effectively blocks ferroptosis in vitro and impairs ferroptosis-associated cancer immunotherapy in vivo. Our results identify PKCßII as a sensor of lipid peroxidation, and the lipid peroxidation-PKCßII-ACSL4 positive-feedback axis may provide potential targets for ferroptosis-associated disease treatment.

Ferroptosis and metabolic dysfunction-associated fatty liver disease: Is there a link?

Non-alcoholic fatty liver disease (NAFLD), recently re-defined and re-classified as metabolic dysfunction-associated fatty liver disease (MAFLD), has become increasingly prevalent and emerged as a public health problem worldwide. To date, the precise pathogenic mechanisms underpinning MAFLD are not entirely understood, and there is no effective pharmacological therapy for NAFLD/MAFLD. As a newly discovered form of iron-dependent programmed cell death, ferroptosis can be involved in the development and progression of various chronic diseases, but the pathogenic connections and mechanisms that link MAFLD and ferroptosis have not been fully elucidated. The main characteristics of ferroptosis are the accumulation of lipid peroxides and reactive oxygen species. In this brief narrative review, the mechanisms of ferroptosis and its putative pathogenic role in MAFLD are discussed to highlight potential new research directions and ideas for the prevention and treatment of MAFLD.

Monascuslactams A-D, cytotoxic γ-lactams from marine fungus Monascus albidus BB3.

Amino acid-directed strategy becomes an efficient way to explore the alkaloids' biosynthetic potential of marine fungi. The metabolites of marine fungus Monascus albidus BB3 were regulated obviously when cultured in GPY medium supplemented with L-tryptophan, L-phenylalanine, D,L-methionine, L-threonine, L-lysine, L-serine and L-valine. Four new γ-lactams, monascuslactams A-D (1-4), together with two known compounds pulchellalactam (5) and O-acetylperlolyrine (6) were obtained. Their structures were determined by comprehensive analysis on the 1 D and 2 D NMR, HRESIMS, UV and IR data, and their absolute configurations were assigned by the experimental and calculated ECD data analysis. Compounds 3, 4 and 6 showed moderate cytotoxicity against human cancer cell lines SUNE1, HepG2, QGY7701, GLC82, HCT116 and MDA-MB-231.

SARS-CoV-2 vulnerability and COVID-19 severity association with ABO blood group / 中国输血杂志

Studies have shown that ABO blood group is related to the susceptibility and disease progression of SARS-CoV-2 infection, and most studies indicated that group O individuals were less likely to get infected while group A conferred a higher susceptibility to infection and propensity to severe disease. ABO blood group antigens are oligosaccharides expressed on red cells and other tissues. People with different ABO blood type have different susceptibility to a variety of pathogens, including SARS-CoV-2. There are several hypotheses to explain the differences in SARS-CoV-2 infection between ABO blood group individuals. Firstly, anti-A and/or anti-B antibodies could bind to corresponding antigens on the viral envelope and contribute to viral neutralization, thereby preventing target cell from being infected. The SARS-CoV-2 virus and SARS-CoV spike (S) proteins may be bound by anti-A isoagglutinin, which may block interactions between virus and angiotensin-converting-enzyme-2-receptor, thereby preventing entering into lung epithelial cells. Secondly, the receptor binding domain (RBD) of S protein domain can bind to antigen A expressed in respiratory epithelium and promote its infection to respiratory epithelial cells. In conclusion, most studies indicated that group O may be associated with a lower risk of SARS-CoV-2 infection while group A with a higher risk along with severe disease, and the related mechanism needs to be further studied.

Prediction of trends for fine-scale spread of Oncomelania hupensis in Shanghai Municipality based on supervised machine learning models / 中国血吸虫病防治杂志

OBJECTIVE@#To predict the trends for fine-scale spread of Oncomelania hupensis based on supervised machine learning models in Shanghai Municipality, so as to provide insights into precision O. hupensis snail control.@*METHODS@#Based on 2016 O. hupensis snail survey data in Shanghai Municipality and climatic, geographical, vegetation and socioeconomic data relating to O. hupensis snail distribution, seven supervised machine learning models were created to predict the risk of snail spread in Shanghai, including decision tree, random forest, generalized boosted model, support vector machine, naive Bayes, k-nearest neighbor and C5.0. The performance of seven models for predicting snail spread was evaluated with the area under the receiver operating characteristic curve (AUC), F1-score and accuracy, and optimal models were selected to identify the environmental variables affecting snail spread and predict the areas at risk of snail spread in Shanghai Municipality.@*RESULTS@#Seven supervised machine learning models were successfully created to predict the risk of snail spread in Shanghai Municipality, and random forest (AUC = 0.901, F1-score = 0.840, ACC = 0.797) and generalized boosted model (AUC= 0.889, F1-score = 0.869, ACC = 0.835) showed higher predictive performance than other models. Random forest analysis showed that the three most important climatic variables contributing to snail spread in Shanghai included aridity (11.87%), ≥ 0 °C annual accumulated temperature (10.19%), moisture index (10.18%) and average annual precipitation (9.86%), the two most important vegetation variables included the vegetation index of the first quarter (8.30%) and vegetation index of the second quarter (7.69%). Snails were more likely to spread at aridity of < 0.87, ≥ 0 °C annual accumulated temperature of 5 550 to 5 675 °C, moisture index of > 39% and average annual precipitation of > 1 180 mm, and with the vegetation index of the first quarter of > 0.4 and the vegetation index of the first quarter of > 0.6. According to the water resource developments and township administrative maps, the areas at risk of snail spread were mainly predicted in 10 townships/subdistricts, covering the Xipian, Dongpian and Tainan sections of southern Shanghai.@*CONCLUSIONS@#Supervised machine learning models are effective to predict the risk of fine-scale O. hupensis snail spread and identify the environmental determinants relating to snail spread. The areas at risk of O. hupensis snail spread are mainly located in southwestern Songjiang District, northwestern Jinshan District and southeastern Qingpu District of Shanghai Municipality.