Type I Photosensitizer Targeting Glycans: Overcoming Biofilm Resistance by Inhibiting the Two-Component System, Quorum Sensing, and Multidrug Efflux.

Stubborn biofilm infections pose serious threats to human health due to the persistence, recurrence, and dramatically magnified antibiotic resistance. Photodynamic therapy has emerged as a promising approach to combat biofilm. Nevertheless, how to inhibit the bacterial signal transduction system and the efflux pump to conquer biofilm recurrence and resistance remains a challenging and unaddressed issue. Herein, a boric acid-functionalized lipophilic cationic type I photosensitizer, ACR-DMP, is developed, which efficiently generates •OH to overcome the hypoxic microenvironment and photodynamically eradicates methicillin-resistant Staphylococcus aureus (MRSA) and biofilms. Furthermore, it not only alters membrane potential homeostasis and osmotic pressure balance due to its strong binding ability with plasma membrane but also inhibits quorum sensing and the two-component system, reduces virulence factors, and regulates the activity of the drug efflux pump attributed to the glycan-targeting ability, helping to prevent biofilm recurrence and conquer biofilm resistance. In vivo, ACR-DMP successfully obliterates MRSA biofilms attached to implanted medical catheters, alleviates inflammation, and promotes vascularization, thereby combating infections and accelerating wound healing. This work not only provides an efficient strategy to combat stubborn biofilm infections and bacterial multidrug resistance but also offers systematic guidance for the rational design of next-generation advanced antimicrobial materials.

A Membrane-Targeted Photosensitizer Prevents Drug Resistance and Induces Immune Response in Treating Candidiasis.

Candida albicans (C. albicans), a ubiquitous polymorphic fungus in humans, causes different types of candidiasis, including oral candidiasis (OC) and vulvovaginal candidiasis (VVC), which are physically and mentally concerning and financially costly. Thus, developing alternative antifungals that prevent drug resistance and induce immunity to eliminate Candida biofilms is crucial. Herein, a novel membrane-targeted aggregation-induced emission (AIE) photosensitizer (PS), TBTCP-QY, is developed for highly efficient photodynamic therapy (PDT) of candidiasis. TBTCP-QY has a high molar absorption coefficient and an excellent ability to generate 1 O2 and •OH, entering the interior of biofilms due to its high permeability. Furthermore, TBTCP-QY can efficiently inhibit biofilm formation by suppressing the expression of genes related to the adhesion (ALS3, EAP1, and HWP1), invasion (SAP1 and SAP2), and drug resistance (MDR1) of C. albicans, which is also advantageous for eliminating potential fungal resistance to treat clinical infectious diseases. TBTCP-QY-mediated PDT efficiently targets OC and VVC in vivo in a mouse model, induces immune response, relieves inflammation, and accelerates the healing of mucosal defects to combat infections caused by clinically isolated fluconazole-resistant strains. Moreover, TBTCP-QY demonstrates excellent biocompatibility, suggesting its potential applications in the clinical treatment of OC and VVC.

Colorimetric and ratiometric supramolecular AIE fluorescent probe for the on-site monitoring of fipronil.

The overuse of fipronil (FPN, a broad-spectrum insecticide) in agriculture has brought great concerns for environmental pollution and food safety. The development of a rapid, reliable, and portable analytical method for the on-site monitoring of FPN is therefore of great significance but is full of challenge. Herein, a novel supramolecular probe using human serum albumin (HSA) as the host and an aggregation-induced emission-active fluorescence probe LIQ-TPA-TZ as the guest was developed for the colorimetric and ratiometric detection of FPN, displaying fast response (30 s), high sensitivity (LOD ∼ 0.05 µM), and good selectivity and anti-interference performance. Moreover, portable paper-based test strips could be facilely obtained and utilized for the determination of FPN, showing colorimetric changes from yellow to orange. This supramolecular probe also demonstrated great potential in real applications for choosing the best cleaning method to reduce the residue rate of FPN on apples. This study provides a versatile tool for the fast and real-time analysis of FPN, which greatly benefits the on-site determination of pesticides with the use of simple testing apparatus.

Molecular Charge and Antibacterial Performance Relationships of Aggregation-Induced Emission Photosensitizers.

Bacterial infections remain a major cause of morbidity worldwide due to drug resistance of pathogenic bacteria. Photodynamic therapy (PDT) has emerged as a promising approach to overcome this drug resistance. However, existing photosensitizers (PSs) are broad-spectrum antibacterial agents that dysregulate the microflora balance resulting in undesirable side effects. Herein, we synthesized a series of aggregation-induced emission (AIE)-active PSs with a lipophilic cationic AIE core with varying charges, named TBTCP and its derivatives. The association of the difference in their molecular charge with the antibacterial effects was systemically investigated. Among the derivatives presented, TBTCP-SF with the electronegative sulfonate group nulled its ability to bind to and ablate Gram-positive (G+) or Gram-negative (G-) bacteria. TBTCP-QY modified by electropositive quaternary ammonium facilitated binding and augmented the photodynamic antibacterial activity for both G+ and G- bacteria. TBTCP-PEG with hydrophilic neutral ligands selectively bound and inactivated G+ bacteria. Under white-light illumination, TBTCP-PEG ablated 99.9% methicillin-resistant Staphylococcus aureus (MRSA) and promoted wound healing in MRSA-infected mice, eliminating MRSA infection both in vitro and in vivo. Our work provides unprecedented insight into the utility of AIE-active PSs for highly targeted and efficient photodynamic ablation of either G+ or G- bacteria that can be translated to next-generation antibacterial materials.

A human VEGF magnetic molecularly imprinted polymer for drug-free anti-angiogenesis and photothermal therapy of tumors.

A magnetic molecularly imprinted polymer was developed with an epitope peptide of human VEGF as a template via an epitope blotting technique. As a drug-free agent, the nanoparticles can significantly suppress the proliferation of tumor cells by integrating anti-angiogenesis and photothermotherapy. This work provides a successful example of the design of multimodal antineoplastic drugs.

Engineered Phage with Aggregation-Induced Emission Photosensitizer in Cocktail Therapy against Sepsis.

Sepsis, a widely recognized disease, is characterized by multiple pathogen infections. Therefore, it is imperative to develop methods that can efficiently identify and neutralize pathogen species. Phage cocktail therapy utilizes the host specificity of phages to adapt to infect resistant bacteria. However, its low sterilization stability efficiency and lack of imaging units seriously restrict its application. Here, a novel strategy combining the aggregation-induced emission photosensitizer (AIE-PS) TBTCP-PMB with phages through a nucleophilic substitution reaction between benzyl bromide and sulfhydryl groups to remove pathogenic bacteria for sepsis treatment is proposed. This strategy retains the phage's host specificity while possessing AIE-PS characteristics with a fluorescence imaging function and reactive oxygen species (ROS) for detecting and sterilizing bacteria. This synergetic strategy combining phage cocktail therapy and photodynamic therapy (PDT) shows a strong "1 + 1 > 2" bactericidal efficacy and superior performance in sepsis mouse models with good biocompatibility. Furthermore, the strategy can quickly diagnose blood infections of clinical blood samples. This simple and accurate strategy provides a promising therapeutic platform for rapid pathogen detection and point-of-care diagnosis. Moreover, it presents a new method for expanding the library of antibacterial drugs to develop new strain identification and improve infectious disease treatment, thereby demonstrating strong translational potential.

The locoregional adiponectin and its synergistic antitumor effect with HIF-1α blockade in TSCC.

Adiponectin (APN) is a kind of endogenous anti-tumor adipocytokine, which exerts its function by binding to its receptors (AdipoR1 and AdipoR2). However, hyperadiponectinemia is found in some pathophysiological processes without significant protective effect, which indicates the existence of APN resistance. Here, we aimed to investigate the locoregional expression of APN in tongue squamous cell carcinoma (TSCC) tissues, and to explore the potential regulatory mechanism of APN resistance under hypoxia. Consequently, we found that the protein expression of APN and AdipoR1, but not AdipoR2, was upregulated in the early stage of TSCC and after hypoxic treatment ex vivo and in vitro. Knockdown of HIF-1α decreased the level of APN and AdipoR1, and simultaneously, HIF-1α was identified as transcriptor of the APN. Intriguingly, a regenerative feedback of HIF-1α was unexpectedly detected after application of recombinant globular APN (gAPN), which most likely contributed to the APN resistance. Furthermore, HIF-1α blockade combined with gAPN has a prominent synergistic antitumor effect, which suggested an effective amelioration in APN resistance. In all, our study revealed the possible mechanism of APN resistance under hypoxia and provides a promising strategy of bi-target treatment with APN and HIF-1α for TSCC therapy.

Synergistically Enhancing Tumor Chemotherapy Using an Aggregation-Induced Emission Photosensitizer on Covalently Conjugated Molecularly Imprinted Polymer Nanoparticles.

Due to the polygenic and heterogeneous nature of the tumorigenesis process, traditional chemotherapy is far from desirable. Fabricating multifunctional nanoplatforms integrating photodynamic effect can synergistically enhance chemotherapy because they can make the cancer cells much sensitive to chemotherapeutics. However, how to assemble different units in nanoplatforms and minimize side effects caused by chemodrugs and photosensitizers (PSs) still needs to be explored. Herein, a nanoplatform CPP/PS-MIP@DOX is developed using a simultaneously covalently conjugated new aggregation-induced emission (AIE) PS and a cell-penetrating peptide (CPP) on the surface of silica-based molecularly imprinted polymer (MIP) nanoparticles, prepared with doxorubicin (DOX) as the template in the water system via a sol-gel technique. CPP/PS-MIP@DOX has good biocompatibility, high DOX-loading ability, promoted cellular uptake, and sustained and pH-sensitive drug release capability. Furthermore, it can efficiently penetrate into tumor tissue, accurately home to, and accumulate at the tumor site. As a result, a better efficacy with lower cytotoxicity is achieved with a smaller dosage of DOX by utilizing either the photodynamic effect or unique characteristics of the MIP. It is the first nanoplatform fabricated by chemically conjugating AIE PSs directly on the surface of the scaffold via the surface-decorated strategy and successfully applied in cancer therapy. This work provides an effective strategy by constructing AIE PS-based cancer nanomedicines with MIPs as scaffolds.

A novel furo[3,2-c]pyridine-based AIE photosensitizer for specific imaging and photodynamic ablation of Gram-positive bacteria.

An Rh-catalyzed tandem reaction was performed to construct an AIE-active furo[2,3-c]pyridine-based photosensitizer, named LIQ-TF. LIQ-TF showed near-infrared emission with high quantum yield, and high 1O2 and ˙OH generation efficiency, and could be used for specific imaging and photodynamic ablation of Gram-positive bacteria in vitro and in vivo, showing great potential for combating multiple drug-resistant bacteria.

Simultaneous Photodynamic Eradication of Tooth Biofilm and Tooth Whitening with an Aggregation-Induced Emission Luminogen.

Dental caries is among the most prevalent dental diseases globally, which arises from the formation of microbial biofilm on teeth. Besides, tooth whitening represents one of the fastest-growing areas of cosmetic dentistry. It will thus be great if tooth biofilm eradication can be combined with tooth whitening. Herein, a highly efficient photodynamic dental therapy strategy is reported for tooth biofilm eradication and tooth discoloration by employing a photosensitizer (DTTPB) with aggregation-induced emission characteristics. DTTPB can efficiently inactivate S. mutans, and inhibit biofilm formation by suppressing the expression of genes associated with extracellular polymeric substance synthesis, bacterial adhesion, and superoxide reduction. Its inhibition performance can be further enhanced through combined treatment with chlorhexidine. Besides, DTTPB exhibits an excellent tooth-discoloration effect on both colored saliva-coated hydroxyapatite and clinical teeth, with short treatment time (less than 1 h), better tooth-whitening performance than 30% hydrogen peroxide, and almost no damage to the teeth. DTTPB also demonstrates excellent biocompatibility with neglectable hemolysis effect on mouse red blood cells and almost no killing effect on mammalian cells, which enables its potential applications for simultaneous tooth biofilm eradication and tooth whitening in clinical dentistry.

Metabolomics analysis of human plasma reveals decreased production of trimethylamine N-oxide retards the progression of chronic kidney disease.

BACKGROUND AND PURPOSE: Chronic kidney disease (CKD) is a global public health problem and one of the leading causes of all-cause mortality. However, the pathogenic mechanisms and intervention methods for CKD progression are not fully understood. EXPERIMENTAL APPROACH: Plasma from patients with uraemia and from healthy controls (n = 30 per group) was analysed with LC-MS/MS-based non-targeted metabolomics to identify potential markers of uraemia. These potential markers were validated in the same cohort and a second cohort (n = 195) by quantitative analysis of the markers, using LC-MS/MS. The most promising marker was identified by correlation analysis and further validated using HK-2 cells and mouse models. KEY RESULTS: Trimethylamine N-oxide (TMAO) was identified as a promising marker among the 18 potential markers found in the first cohort, and it was optimally correlated with renal function of CKD patients in the second cohort. Treatment of HK-2 cells with TMAO decreased cell viability and up-regulated expression of α-smooth muscle actin. In mice, a TMAO-containing diet decreased kidney mass and increased protein expression of α-smooth muscle actin. Also, control of TMAO production by inhibiting its biosynthetic pathway with 3,3-dimethyl-1-butanol or disrupting gut microbiota function with an antibiotic cocktail, attenuated renal injury in a murine model of CKD. CONCLUSION AND IMPLICATIONS: Our data show that decreased TMAO production could be a new strategy to attenuate the progression of renal injury in CKD.

Copper mediated molecularly imprinted polymers for fast recognizing tylosin.

As one kind of artificial antibody, molecularly imprinted polymers (MIPs) has been widely used to separate/enrich target molecules from samples with the complex matrix prior to instrumental analysis. In this work, one novel copper mediated magnetic MIPs was developed towards anti-bacteria macrolide antibiotic tylosin (TYL). The obtained microspheres had a lot of convexities distributed evenly on the surface. And it exhibited high hydrophilicity and superparamagnetic ability, as well as excellent selectivity and specificity. Notably, it only took 5 min to reach the absorption equilibrium for TYL, which made it highly feasible for high-throughput analysis. Furthermore, the microspheres was applied as the magnetic dispersed solid phase extraction material to enrich trace TYL residue in (spiked) human urine. At three spiking levels, mean recoveries are in the range of 76.42-93.72% with relative standard deviations of 2.75-8.24% (n = 3) with HPLC/UV-Vis. The work provided one promising reference to prepare novel solid phase extraction materials for enriching/separating trace component in complex matrix.

A near-infrared AIE fluorescent probe for myelin imaging: From sciatic nerve to the optically cleared brain tissue in 3D.

Myelin, the structure that surrounds and insulates neuronal axons, is an important component of the central nervous system. The visualization of the myelinated fibers in brain tissues can largely facilitate the diagnosis of myelin-related diseases and understand how the brain functions. However, the most widely used fluorescent probes for myelin visualization, such as Vybrant DiD and FluoroMyelin, have strong background staining, low-staining contrast, and low brightness. These drawbacks may originate from their self-quenching properties and greatly limit their applications in three-dimensional (3D) imaging and myelin tracing. Chemical probes for the fluorescence imaging of myelin in 3D, especially in optically cleared tissue, are highly desirable but rarely reported. We herein developed a near-infrared aggregation-induced emission (AIE)-active probe, PM-ML, for high-performance myelin imaging. PM-ML is plasma membrane targeting with good photostability. It could specifically label myelinated fibers in teased sciatic nerves and mouse brain tissues with a high-signal-to-background ratio. PM-ML could be used for 3D visualization of myelin sheaths, myelinated fibers, and fascicles with high-penetration depth. The staining is compatible with different brain tissue-clearing methods, such as ClearT and ClearT2 The utility of PM-ML staining in demyelinating disease studies was demonstrated using the mouse model of multiple sclerosis. Together, this work provides an important tool for high-quality myelin visualization across scales, which may greatly contribute to the study of myelin-related diseases.

Machine Learning to Improve Prognosis Prediction of Early Hepatocellular Carcinoma After Surgical Resection.

BACKGROUND: Improved prognostic prediction is needed to stratify patients with early hepatocellular carcinoma (EHCC) to refine selection of adjuvant therapy. We aimed to develop a machine learning (ML)-based model to predict survival after liver resection for EHCC based on readily available clinical data. METHODS: We analyzed data of surgically resected EHCC (tumor≤5 cm without evidence of extrahepatic disease or major vascular invasion) patients from the Surveillance, Epidemiology, and End Results (SEER) Program to train and internally validate a gradient-boosting ML model to predict disease-specific survival (DSS). We externally tested the ML model using data from 2 Chinese institutions. Patients treated with resection were matched by propensity score to those treated with transplantation in the SEER-Medicare database. RESULTS: A total of 2778 EHCC patients treated with resection were enrolled, divided into 1899 for training/validation (SEER) and 879 for test (Chinese). The ML model consisted of 8 covariates (age, race, alpha-fetoprotein, tumor size, multifocality, vascular invasion, histological grade and fibrosis score) and predicted DSS with C-Statistics >0.72, better than proposed staging systems across study cohorts. The ML model could stratify 10-year DSS ranging from 70% in low-risk subset to 5% in high-risk subset. Compared with low-risk subset, no remarkable survival benefits were observed in EHCC patients receiving transplantation before and after propensity score matching. CONCLUSION: An ML model trained on a large-scale dataset has good predictive performance at individual scale. Such a model is readily integrated into clinical practice and will be valuable in discussing treatment strategies.

A Membrane-Targeting Photosensitizer with Aggregation-Induced Emission Characteristics for Highly Efficient Photodynamic Combat of Human Coronaviruses.

COVID-19 pandemic, caused by severe acute respiratory syndrome coronavirus 2, has resulted in global social and economic disruption, putting the world economy to the largest global recession since the Great Depression. To control the spread of COVID-19, cutting off the transmission route is a critical step. In this work, the efficient inactivation of human coronavirus with photodynamic therapy (PDT) by employing photosensitizers with aggregation-induced emission characteristics (DTTPB) is reported. DTTPB is designed to bear a hydrophilic head and two hydrophobic tails, mimicking the structure of phospholipids on biological membranes. DTTPB demonstrates a broad absorption band covering the whole visible light range and high molar absorptivity, as well as excellent reactive oxygen species sensitizing ability, making it an excellent candidate for PDT. Besides, DTTPB can target membrane structure, and bind to the envelope of human coronaviruses. Upon light irradiation, DTTPB demonstrates highly effective antiviral behavior: human coronavirus treated with DTTPB and white-light irradiation can be efficiently inactivated with complete loss of infectivity, as revealed by the significant decrease of virus RNA and proteins in host cells. Thus, DTTPB sensitized PDT can efficiently prevent the infection and the spread of human coronavirus, which provides a new avenue for photodynamic combating of COVID-19.

Simple synthesis of multifunctional photosensitizers for mitochondrial and bacterial imaging and photodynamic anticancer and antibacterial therapy.

Photosensitizers (PSs), a critical drug administered for successful photodynamic therapy (PDT), have been well researched regarding their anticancer or bactericidal capability with high precision and low invasiveness. Although traditional PSs have been explored either in photodynamic anticancer or in antibiosis, they usually require synthesis with multiple steps, harsh synthetic conditions, and a complicated purification process for a single targeted product. Therefore, developing new multifunctional PSs with a simple synthesis and reactant flexibility which combine mitochondrial and bacterial imaging, efficient photodynamic anticancer and antibacterial effects is of the utmost urgency and of great importance for clinical applications. Herein, a large structural investigation of isoquinolinium-based PSs synthesized by a simple Rh-catalysed annulation reaction with high yields is presented. These lipophilic cationic PSs have a tunable photophysical property. LIQ-6 was found to perform not only as an ideal mitochondria targeting probe but also an effective cancer cell killing PS, and moreover, a tracker for bacterial imaging and ablation. LIQ-6 can be used to image a wide range of cancer cells and to monitor the photo-induced cell apoptosis, and simultaneously, it can also image and be a photodynamic germicide for both Gram-positive and Gram-negative bacteria. Furthermore, LIQ-6 shows great effectiveness in the wound healing process, showing its ability to be an ideal PS in vivo as well. This contribution is believed to offer a new platform for the construction of a theragnostic system for future practical applications in biology and biomedicine.

Development and Validation of Pre- and Post-Operative Models to Predict Recurrence After Resection of Solitary Hepatocellular Carcinoma: A Multi-Institutional Study.

BACKGROUND: The ideal candidates for resection are patients with solitary hepatocellular carcinoma (HCC); however, postoperative recurrence rate remains high. We aimed to establish prognostic models to predict HCC recurrence based on readily accessible clinical parameters and multi-institutional databases. PATIENTS AND METHODS: A total of 485 patients undergoing curative resection for solitary HCC were recruited from two independent institutions and the Cancer Imaging Archive database. We randomly divided the patients into training (n=323) and validation cohorts (n=162). Two models were developed: one using pre-operative and one using pre- and post-operative parameters. Performance of the models was compared with staging systems. RESULTS: Using multivariable analysis, albumin-bilirubin grade, serum alpha-fetoprotein and tumor size were selected into the pre-operative model; albumin-bilirubin grade, serum alpha-fetoprotein, tumor size, microvascular invasion and cirrhosis were selected into the postoperative model. The two models exhibited better discriminative ability (concordance index: 0.673-0.728) and lower prediction error (integrated Brier score: 0.169-0.188) than currently used staging systems for predicting recurrence in both cohorts. Both models stratified patients into low- and high-risk subgroups of recurrence with distinct recurrence patterns. CONCLUSION: The two models with corresponding user-friendly calculators are useful tools to predict recurrence before and after resection that may facilitate individualized management of solitary HCC.

A simple yet effective AIE-based fluorescent nano-thermometer for temperature mapping in living cells using fluorescence lifetime imaging microscopy.

We designed and synthesized a novel nano-thermometer using aggregation-induced-emission (AIE) dye as the reporter and household butter as the matrix. This temperature nanosensor showed decreased fluorescence intensities (∼2%/°C) and shorter fluorescence lifetimes (∼0.11 ns/°C) upon increasing the environmental temperature in the physiological temperature range. Such fluorescence responses were reversible and independent of the environmental pH and ionic strength. The application of these nano-thermometers in temperature sensing in living cells using fluorescence lifetime imaging microscopy (FLIM) was also demonstrated. To the best of our knowledge, this is the first example of AIE-based nano-thermometer for temperature sensing in living cells. This work also provides us with a simple and low-cost method for rapid fabrication of an effective nanosensor based on AIE mechanism.

A Small-Molecule AIE Chromosome Periphery Probe for Cytogenetic Studies.

The chromosome periphery (CP) is a complex network that covers the outer surface of chromosomes. It acts as a carrier of nucleolar components, helps maintain chromosome structure, and plays an important role in mitosis. Current methods for fluorescence imaging of CP largely rely on immunostaining. We herein report a small-molecule fluorescent probe, ID-IQ, which possesses aggregation-induced emission (AIE) property, for CP imaging. By labelling the CP, ID-IQ sharply highlighted the chromosome boundaries, which enabled rapid segmentation of touching and overlapping chromosomes, direct identification of the centromere, and clear visualization of chromosome morphology. ID-IQ staining was also compatible with fluorescence in situ hybridization and could assist the precise location of the gene in designated chromosome. Altogether, this study provides a versatile cytogenetic tool for improved chromosome analysis, which greatly benefits the clinical diagnostic testing and genomic research.

Radiomic Features at Contrast-enhanced CT Predict Recurrence in Early Stage Hepatocellular Carcinoma: A Multi-Institutional Study.

Background Early stage hepatocellular carcinoma (HCC) is the ideal candidate for resection in patients with preserved liver function; however, cancer will recur in half of these patients and no reliable prognostic tool has been established. Purpose To investigate the effectiveness of radiomic features in predicting tumor recurrence after resection of early stage HCC. Materials and Methods In total, 295 patients (median age, 58 years; interquartile range, 50-65 years; 221 men) who underwent contrast material-enhanced CT and curative resection for early stage HCC that met the Milan criteria between February 2009 and December 2016 were retrospectively recruited from three independent institutions. Follow-up consisted of serum α-fetoprotein level, liver function tests, and dynamic imaging examinations every 3 months during the first 2 years and then every 6 months thereafter. In the development cohort of 177 patients from institution 1, recurrence-related radiomic features were computationally extracted from the tumor and its periphery and a radiomics signature was built with least absolute shrinkage and selection operator regression. Two models, one integrating preoperative and one integrating pre- and postoperative variables, were created by using multivariable Cox regression analysis. An independent external cohort of 118 patients from institutions 2 and 3 was used to validate the proposed models. Results The preoperative model integrated radiomics signature with serum α-fetoprotein level and tumor number; the postoperative model incorporated microvascular invasion and satellite nodules into the above-mentioned predictors. In both study cohorts, two radiomics-based models provided better predictive performance (concordance index ≥0.77, P < .05 for all), lower prediction error (integrated Brier score ≤0.14), and larger net benefits, as determined by means of decision curve analysis, than rival models without radiomics and widely adopted staging systems. The radiomics-based models gave three risk strata with high, intermediate, or low risk of recurrence and distinct profiles of recurrent tumor number. Conclusion The proposed radiomics models with pre- and postresection features helped predict tumor recurrence for early stage hepatocellular carcinoma. © RSNA, 2020 Online supplemental material is available for this article.