Machine learning methods, databases and tools for drug combination prediction.

Combination therapy has shown an obvious efficacy on complex diseases and can greatly reduce the development of drug resistance. However, even with high-throughput screens, experimental methods are insufficient to explore novel drug combinations. In order to reduce the search space of drug combinations, there is an urgent need to develop more efficient computational methods to predict novel drug combinations. In recent decades, more and more machine learning (ML) algorithms have been applied to improve the predictive performance. The object of this study is to introduce and discuss the recent applications of ML methods and the widely used databases in drug combination prediction. In this study, we first describe the concept and controversy of synergism between drug combinations. Then, we investigate various publicly available data resources and tools for prediction tasks. Next, ML methods including classic ML and deep learning methods applied in drug combination prediction are introduced. Finally, we summarize the challenges to ML methods in prediction tasks and provide a discussion on future work.

Computational methods, databases and tools for synthetic lethality prediction.

Synthetic lethality (SL) occurs between two genes when the inactivation of either gene alone has no effect on cell survival but the inactivation of both genes results in cell death. SL-based therapy has become one of the most promising targeted cancer therapies in the last decade as PARP inhibitors achieve great success in the clinic. The key point to exploiting SL-based cancer therapy is the identification of robust SL pairs. Although many wet-lab-based methods have been developed to screen SL pairs, known SL pairs are less than 0.1% of all potential pairs due to large number of human gene combinations. Computational prediction methods complement wet-lab-based methods to effectively reduce the search space of SL pairs. In this paper, we review the recent applications of computational methods and commonly used databases for SL prediction. First, we introduce the concept of SL and its screening methods. Second, various SL-related data resources are summarized. Then, computational methods including statistical-based methods, network-based methods, classical machine learning methods and deep learning methods for SL prediction are summarized. In particular, we elaborate on the negative sampling methods applied in these models. Next, representative tools for SL prediction are introduced. Finally, the challenges and future work for SL prediction are discussed.

Enhanced Photoelectric Properties of CsPbBr3 by SiO2 and TiO2 Bilayer Heterostructures.

CsPbBr3/SiO2 heterostructures were synthesized by the hydrolysis reaction of a mixture of CsPbBr3 nanocrystals (NCs) and (3-aminopropyl)triethoxysilane (APTS) in air. Compared with CsPbBr3 NCs, the CsPbBr3/SiO2 heterostructures exhibit stronger photoluminescence (PL) intensity, longer lifetime of PL (∼40.5 ns), and higher PL-quantum yield (PLQY, ∼86%). The carrier dynamics of CsPbBr3/SiO2 was detected by the transient absorption (TA) spectrum. The experimental results show that SiO2 passivates the surface traps of CsPbBr3 NCs and enhances the PL intensity. However, photoelectrochemical impedance spectra (PEIS) demonstrate that the impedance of CsPbBr3/SiO2 is higher than that of CsPbBr3 NCs, which reduces carrier transport and extraction. Because the application of CsPbBr3/SiO2 in optoelectronics is limited, CsPbBr3/SiO2/TiO2 heterostructures were synthesized by the further reaction of tetrabutyl titanate (TBT). The TiO2 coating can reduce the impedance of the CsPbBr3/SiO2. Importantly, ∼68% of the PL intensity of CsPbBr3/SiO2 is retained. Compared with CsPbBr3/SiO2 and CsPbBr3 NCs, the CsPbBr3/SiO2/TiO2 demonstrates faster carrier transport (κct = 2.4 × 109 s-1) and higher photocurrent density (J = 76 nA cm-2). In addition, CsPbBr3/SiO2/TiO2 shows good stability under (ultraviolet) UV irradiation, along with water stability and thermal stability. Therefore, the double protection approach can enhance the stability of CsPbBr3 NCs and tune the optoelectronic properties of CsPbBr3 NCs.

Texture-Based Classification of Fetal Growth Restriction From Intrauterine Neurosonographic Image.

OBJECTIVE: Fetal growth restriction (FGR) is a condition where fetuses fail to reach their genetic potential for growth, posing a significant health challenge for newborns. The aim of this research was to explore the efficacy of texture-based analysis of neurosonographic images in identifying FGR in fetuses, which may provide a promising tool for early assessment of FGR. METHODS: A retrospective analysis collected 100 intrauterine neurosonographic images from 50 FGR and 50 gestational age-appropriate fetuses. Using MaZda software, approximately 300 texture features were extracted from occipital white matter (OWM) and cerebellum of intrauterine neurosonographic images, respectively. Then 10 optimal features were separately selected by 3 algorithms, including the Fisher coefficient method, the method of minimizing classification error probability and average correlation coefficients, and the mutual information coefficient method. Further, the 10 statistically most significant features were selected from these sets to form the mixed feature set. After nonlinear discriminant analysis was performed to reduce feature dimensionality, the artificial neural network (ANN) classifier was conducted, respectively. RESULTS: For OWM and cerebellum, a total of 11 and 14 statistically significant features were selected. When the mixed feature sets of OWM and cerebellum were applied to ANN classifier, classification accuracy were 90.00% (κ = 0.800; P < .001) and 93.00% (κ = 0.860; P < .001), and the receiver operating characteristic curve for identifying FGR showed an area under the curve of 0.82 and 0.87. CONCLUSIONS: Texture analysis of fetal intrauterine neurosonographic images is a feasible and noninvasive strategy for evaluating FGR fetuses.

New insights into the degradation and detoxification of methylene blue using heterogeneous-Fenton catalyzed by sustainable siderite.

The huge application of synthetic dyes caused a severe impact in the environment. In the present study, a physico-chemical strategy of heterogeneous-Fenton catalyzed by the natural ferrous ore has been established for toxic chemical degradation, of which the complex and high-expense repetitive pH adjustment procedures were escaping. And this natural heterogeneous catalyst also could be recycled and sustainable for toxic substances treatment involved in synergetic adsorption and oxidation. The siderite, served as an adsorbent and catalyst for the degradation of methylene blue (MB). Siderite exhibited a better adsorption capacity with a saturated adsorption capacity of ∼11.08 mg/g. Batch adsorption experiments have verified that adsorption rate and adsorption equilibrium followed pseudo-second-order rate model and Langmuir isotherm equation, respectively. The combination with H2O2, showed significant enhancement of MB degradation without any pH adjustment. The effect of siderite dosage, H2O2 dosage, MB concentration, initial pH, and reaction temperature on MB degradation was investigated, which also has indicated the excellent catalytic performance of siderite. About 99.71% of MB was degraded in 480 min with initial pH of 7.0, reaction temperature of 25 °C, siderite, and H2O2 dosage of 2.5 g/L and 122.38 mM, respectively. It was found that siderite could be reused and remained high degradation efficiency on MB after 5 times reutilization, which also could demonstrate the sustainable and effective process to degrade organic pollution. The generation of reactive species including ·OH and O2·- have been confirmed based on scavenger test and electron spin resonance (ESR) analysis, which was dominated by heterogeneous reaction. The possible degradation mechanisms of MB have been predicted based on spectrum scanning and GC-MS analysis. Moreover, acute toxicity assessment with marine photobacterium Vibrio fisheri was conducted to investigate the toxicity change in the adsorption/oxidation coupled process. This sustainable heterogeneous-Fenton technology has been verified as a promising and applicable process for toxic organic chemicals removal due to effective mineralization and detoxification assisted with the natural ore mineral through the simple operation and mild condtions.

DEML: Drug Synergy and Interaction Prediction Using Ensemble-Based Multi-Task Learning.

Synergistic drug combinations have demonstrated effective therapeutic effects in cancer treatment. Deep learning methods accelerate identification of novel drug combinations by reducing the search space. However, potential adverse drug-drug interactions (DDIs), which may increase the risks for combination therapy, cannot be detected by existing computational synergy prediction methods. We propose DEML, an ensemble-based multi-task neural network, for the simultaneous optimization of five synergy regression prediction tasks, synergy classification, and DDI classification tasks. DEML uses chemical and transcriptomics information as inputs. DEML adapts the novel hybrid ensemble layer structure to construct higher order representation using different perspectives. The task-specific fusion layer of DEML joins representations for each task using a gating mechanism. For the Loewe synergy prediction task, DEML overperforms the state-of-the-art synergy prediction method with an improvement of 7.8% and 13.2% for the root mean squared error and the R2 correlation coefficient. Owing to soft parameter sharing and ensemble learning, DEML alleviates the multi-task learning 'seesaw effect' problem and shows no performance loss on other tasks. DEML has a superior ability to predict drug pairs with high confidence and less adverse DDIs. DEML provides a promising way to guideline novel combination therapy strategies for cancer treatment.

Efficacy and safety of a selective URAT1 inhibitor SHR4640 in Chinese subjects with hyperuricaemia: a randomized controlled phase II study.

OBJECTIVE: To evaluate the efficacy and safety of SHR4640, a highly selective urate transporter 1 inhibitor, in Chinese subjects with hyperuricaemia. METHODS: This was a randomized double-blind dose-ranging phase II study. Subjects whose serum uric acid (sUA) levels were ≥480 µmol/l with gout, ≥480 µmol/l without gout but with comorbidities, or ≥540 µmol/l were enrolled. Subjects were randomly assigned (1:1:1:1:1) to receive once daily 2.5 mg, 5 mg, 10 mg of SHR4640, 50 mg of benzbromarone or placebo, respectively. The primary end point was the proportion of subjects who achieved target sUA level of ≤360 µmol/l at week 5. RESULTS: 99.5% of subjects (n = 197) were male and 95.9% of subjects had gout history. The proportions of subjects who achieved target sUA at week 5 were 32.5%, 72.5% and 61.5% in the 5 mg, 10 mg SHR4640 and benzbromarone groups, respectively, significantly higher than the placebo group (0%; P < 0.05 for 5 mg and 10 mg SHR4640 group). The sUA was reduced by 32.7%, 46.8% and 41.8% at week 5 with 5 mg, 10 mg SHR4640 and benzbromarone, respectively, vs placebo (5.9%; P < 0.001 for each comparison). The incidences of gout flares requiring intervention were similar among all groups. Occurrences of treatment-emergent adverse events (TEAEs) were comparable across all groups, and serious TEAEs were not reported. CONCLUSIONS: The present study indicated a superior sUA-lowering effect and well tolerated safety profile after 5-week treatment with once-daily 5 mg/10 mg of SHR4640 as compared with placebo in Chinese subjects with hyperuricaemia. TRIAL REGISTRATION: ClinicalTrials.gov number, NCT03185793.

H2-oxidation driven by its behavior of losing an electron over B-doped TiO2 under UV irradiation.

In this work, TiO2 was modified by doping the electron-deficient B element, and then the gas-sensing response of B-TiO2 to H2 under UV irradiation at room temperature in a N2 atmosphere and the oxidation of H2 over B-TiO2 under corresponding conditions were tested. It was found that H2 would accept an electron when adsorbed on the TiO2 surface, while H2 would donate an electron when adsorbed on the B-TiO2 surface. Correspondingly, H2 could not be oxidized over TiO2, but could be oxidized over B-TiO2. This indicated that the oxidation of H2 was dependent on the electron-transfer behavior between H2 and the surface of TiO2 or B-TiO2. Based on the relevant characterization results, it was proposed that H2 could accept an electron from TiO2 due to the higher Fermi level of TiO2, while H2 could donate an electron to B-TiO2 due to the lower Fermi level of B-TiO2 induced by doping B. This indicated that the electron-transfer behavior between H2 and TiO2 could be changed by adjusting the Fermi level of TiO2, while the electron-transfer behavior would further affect the photocatalytic activity of oxidizing H2. This result shows that the doable H2 photocatalytic oxidation in thermodynamics can be controlled by a kinetics factor (H2 losing-an-electron behavior). This work can be applied to provide an understanding of the photocatalytic oxidation behavior of other reactants over semiconductor materials.

Treatment of Synthetic Ammonium Sulfate Wastewater by Mixed Culture of Chlorella pyrenoidosa and Enriched Nitrobacteria.

Ammonium sulfate wastewater can cause eutrophication and black odor of water body. Although ammonia nitrogen can be used as nutrient of microalgae, high ammonia nitrogen levels could inhibit the growth of microalgae. Nitrobacteria can transform ammonia nitrogen into nitrate nitrogen. In this study, mono Chlorella pyrenoidosa culture (mono-C.py), synchronous mixed culture (mixed-a), and asynchronous mixed culture (mixed-b) systems were examined for their ability to treat ammonium sulfate wastewater. Nitrogen removal rate of mixed-b at the end of culture (52.96%) was higher than that of the mono-C.py (46.37%) and the mixed-a (39.11%). Higher total suspended solid concentration (2.40 g/L), crude protein yield (0.76 g/L), and heating value yield (35.73 kJ/L) were obtained in mixed-b, meanwhile with excellent settlement performance (91.43 ± 0.51%). Mechanism analysis of settlement showed that the relative abundance of floc-forming-related bacteria Sphingopyxis and Acidovorax were increased generally, while nitrification/denitrifying members were decreased in mixed-b along with the culture proceeding.

Low-dose apatinib and tegafur-gimeracil-oteracil as palliative treatment in recurrent cervical cancer patients with poor performance status: A case series.

WHAT IS KNOWN AND OBJECTIVE: A large proportion of recurrent cervical cancer (RCC) patients present with poor performance status (PS) after comprehensive treatments, which usually prevents them from opting for clinical trials. We retrospectively analysed the effect and safety of low-dose apatinib and tegafur-gimeracil-oteracil (TGO) in the treatment of these patients. CASE SUMMARY: Six patients treated with low-dose apatinib and TGO showed a disease control rate of 83.3% and grade 1-2 adverse events (AEs). WHAT IS NEW AND CONCLUSION: This case series indicates that low-dose apatinib and TGO could be considered as palliative therapy for RCC patients with poor PS.

Engineering Corynebacterium glutamicum Mutants for 3-Methyl-1-butanol Production.

3-Methyl-1-butanol (3MB) is a promising biofuel that can be produced from 2-ketoisocaproate via the common L-leucine biosynthesis pathway. Corynebacterium glutamicum was chosen as a host bacterium because of its strong resistance to isobutanol. In the current study, several strategies were designed to overproduce 3MB in C. glutamicum through a non-fermentation pathway. The engineered C. glutamicum mutant was obtained by silencing the pyruvate dehydrogenase gene complex (aceE) and deleting the lactic dehydrogenase gene (ldh), followed by mutagenesis with diethyl sulfate (DES) and selection with Fmoc-3-4-thiazolyl-L-alanine (FTA). The mutant could produce 659 mg/L of 3MB after 12 h of incubation. To facilitate carbon flux to 3MB biosynthesis, the engineered recombinant was also constructed without branched-chain acid aminotransferase (ilvE) activity by deleting the ilvE gene. This recombinant could produce 697 mg/L of 3MB after 12 h of incubation.

Preparation of hierarchically porous carbon from cellulose as highly efficient adsorbent for the removal of organic dyes from aqueous solutions.

The hierarchically porous carbons were prepared from cellulose by a one-step method and studied as dye-adsorbents using chrysoidine as a model. With the increase of holding temperature, the prepared porous carbons had larger specific surface area and became more effective. Kinetic analysis revealed that adsorption kinetics obeyed the pseudo-second order kinetic model and adsorption equilibrium could reach within 1 h. The studies on isotherm indicated that the adsorption process could be best described by the Langmuir isotherm model and the maximum adsorption capacities was 598.8 mg g-1. Thermodynamic parameters demonstrated that the adsorption process was exothermic and spontaneous. Moreover, the porous carbon could effectively remove some other common dyes. Importantly, the prepared porous carbon was able to remove 98% of chrysoidine from polluted river water and its adsorption efficiency was inappreciably influenced by the water matrix. These results shown that the hierarchically porous carbons synthesized from biomass in this study had a wide application in water treatment.

The mixed culture of microalgae Chlorella pyrenoidosa and yeast Yarrowia lipolytica for microbial biomass production.

Microbial biomass which mostly generated from the microbial processes of bacteria, yeasts, and microalgae is an important resource. Recent concerns in microbial biomass production field, especially microbial lipid production for biofuel, have been focused towards the mixed culture of microalgae and yeast. To more comprehensive understanding of the mixed culture for microbial biomass, mono Chlorella pyrenoidosa, mono Yarrowia lipolytica and the mixed culture were investigated in the present work. Results showed that the mixed culture achieved significantly faster cell propagation of microalga and yeast, smaller individual cell size of yeast and higher relative chlorophyll content of microalga. The mixed culture facilitated the assimilation of carbon and nitrogen and drove the carbon flow to carbohydrate. Besides higher lipid yield (0.77 g/L), higher yields of carbohydrates (1.82 g/L), protein (1.99 g/L) and heating value (114.64 kJ/L) indicated the microbial biomass harvested from the mixed culture have more potential utilization in renewable energy, feedstuff, and chemical industry.

Design and Application of Toxic and Harmful Gas Monitoring System in Fire Fighting.

In recent years, fire accidents in petrochemical plant areas and dangerous goods storage ports in China have shown a trend of frequent occurrence. Toxic and harmful gases are diffused in the scenes of these accidents, which causes great difficulties for fire fighting and rescue operations of fire fighting forces, and consequently, casualties of firefighters often occur. In order to ensure the safety of firefighters in such places, this paper designs a monitoring system of toxic and harmful gases specially used in fire fighting and rescue sites of fire forces, and establishes the transmission network, monitoring terminal and data processing software of the monitoring system of toxic and harmful gases, establishing the danger model of the monitoring area of toxic and harmful gas-monitoring terminal, and the danger model of fire fighters' working area, fusing the field toxic and harmful gas data, terminal positioning data, and field environmental data, designing the data structure of the input data set and the network structure of the RNN cyclic neural network model, and realizing the dynamic early warning of toxic and harmful gases on site.

Effects of low concentrations copper on antioxidant responses, DNA damage and genotoxicity in thick shell mussel Mytilus coruscus.

For the aim to study potential detrimental effects induced by Cu exposure at low, environmentally relevant concentrations, the in vivo activities at different levels of biological organisations of thick shell mussel Mytilus coruscus exposed to two levels of copper were assessed. Cu-induced stresses were evaluated through antioxidant responses, DNA damage and genotoxicity. The results revealed significant higher SOD and CAT activities, and MDA concentration in haemocytes of M. coruscus with Cu exposure at 8 µg/L, while only significant accumulation in CAT activity with Cu exposure at 2 µg/L and no significant changes with SOD activity and MDA concentration at this level of Cu exposure. Copper exposure induced DNA damage as induction of OTM value in a time- and concentration-dependent manner. In addition, copper exposure could significantly induced the expressions of MT-10, Hsp70, Hsp90 and C3. The present results deepen the mussels as a suitable model marine invertebrate species to study potential detrimental effects induced by possible toxicants, in combinations at different levels of biological organisations.

Mixed matrix membranes with fast and selective transport pathways for efficient CO2 separation.

To improve CO2 separation performance, porous carbon nanosheets (PCNs) were used as a filler into a Pebax MH 1657 (Pebax) matrix, fabricating mixed matrix membranes (MMMs). The PCNs exhibited a preferential horizontal orientation within the Pebax matrix because of the extremely large 2D plane and nanoscale thickness of the matrix. Therefore, the micropores of the PCNs provided fast CO2 transport pathways, which led to increased CO2 permeability. The reduced pore size of the PCNs was a consequence of the overlapping of PCNs and the polymer chains penetrating into the pores of the PCNs. The reduction in the pore size of the PCNs improved the CO2/gas selectivity. As a result, the CO2 permeability and CO2/CH4 selectivity of the Pebax membrane with 10 wt% PCNs-loading (Pebax-PCNs-10) were 520 barrer and 51, respectively, for CO2/CH4 mixed-gas. The CO2 permeability and CO2/N2 selectivity of the Pebax-PCNs-10 membrane were 614 barrer and 61, respectively, for CO2/N2 mixed-gas.

iRhom2 deficiency relieves TNF-α associated hepatic dyslipidemia in long-term PM2.5-exposed mice.

Accumulating researches reported that particulate matter (PM2.5) is a risk factor for developing various diseases, including metabolic syndrome. Recently, inactive rhomboid protein 2 (iRhom2) was considered as a necessary modulator for shedding of tumor necrosis factor-α (TNF-α) in immune cells. TNF-α, a major pro-inflammatory cytokine, was linked to various pathogenesis of diseases, including dyslipidemia. Here, wild type (WT) and iRhom2-knockout (iRhom2-/-) mice were used to investigate the effects of iRhom2 on PM2.5-induced hepatic dyslipidemia. The hepatic histology, inflammatory response, glucose tolerance, serum parameters and gene expressions were analyzed. We found that long-term inhalation of PM2.5 resulted in hepatic steatosis. And a significant up-regulation of iRhom2 in liver tissues was observed, accompanied with elevated TNF-α, TNF-α converting enzyme (TACE), TNFα receptor (TNFR)2 and various inflammatory cytokines expressions. Additionally, PM2.5 treatment caused TG and TC accumulation in serum and liver, probably attributed to changes of genes modulating lipid metabolism. Intriguingly, hepatic injury and dyslipidemia were attenuated by iRhom2-/- in mice with PM2.5 challenge. In vitro, iRhom2-knockdwon reduced TNF-α expressions and its associated inflammatory cytokines in Kupffer cells, implying that liver-resident macrophages played an important role in regulating hepatic inflammation and lipid metabolism in cells treated with PM2.5. The findings indicated that long-term PM2.5 exposure caused hepatic steatosis and dyslipidemia through triggering inflammation, which was, at least partly, dependent on iRhom2/TNF-α pathway in liver-resident macrophages.

Effects of low-frequency pulsed electromagnetic fields on plateau frostbite healing in rats.

Plateau frostbite (PF) treatments have remained a clinical challenge because this condition injures tissues in deep layers and affected tissues exhibit unique pathological characteristics. For instance, low-frequency pulsed electromagnetic field (PEMF) can affect tissue restoration and penetrate tissues. Therefore, the effect of PEMF on PF healing should be investigated. This study aimed to evaluate the effects of low-frequency PEMF on PF healing systematically. Ninety-six Sprague-Dawley rats were randomly and equally divided into three groups: normal control, partial thickness plateau frostbite (PTPF), and PTPF with low-frequency PEMF exposure (PTPF + PEMF). PTPF wounds were induced in the dorsum of the rats. The PTPF + PEMF group was exposed to low-frequency PEMF daily. During PF healing, wound microcirculation in each group was monitored through contrast ultrasonography. Wound appearance, histological observation, and wound tensile strength were also evaluated. Results showed that the rate of the microcirculation restoration of the PTPF + PEMF group was nearly 25% faster than that of the PTPF group, and wound appearance suggested that the healing of the PTPF group was slower than that of the PTPF + PEMF group. Histological observation revealed that PEMF accelerated the growth of different deep tissues, as confirmed by tensile strength examination. Low-frequency PEMF could penetrate PF tissues, promote their restoration, and provide a beneficial effect on PF healing. Therefore, this technique may be a potential alternative to treat PF.

Comparative Study of Two Different TiO2 Film Sensors on Response to H2 under UV Light and Room Temperature.

An anatase TiO2 film sensor was prepared by a facile in-situ method on the interdigitated Au electrode deposited on the alumina substrate. The structure, morphology and the optical properties of the in-situ TiO2 film sensor were characterized by X-ray diffraction, Scanning Electron Microscopy, and UV-vis diffuse reflectance spectra. The photo-assisted gas sensitivities of the prepared film towards H2 gas were evaluated at room temperature in N2 and synthetic air atmospheres. As compared to TiO2 film sensor prepared by drop-coating method, this in-situ TiO2 film sensor exhibited a more compact structure composed of uniform TiO2 microspheres as well as a better gas sensitivity towards H2 under UV irradiation, especially in synthetic air. The photo-electrochemical measurements suggest that these improvements may be associated with the efficient charge transfer in the TiO2 interface induced by the TiO2 microsphere structure. This study might offer a feasible approach to develop photo-assisted gas sensors at ambient temperature.

STAT3 inhibitor stattic enhances radiosensitivity in esophageal squamous cell carcinoma.

The radioresistance of esophageal squamous cell carcinoma (ESCC) remains an obstacle for the effective radiotherapy of ESCC. This study aimed to investigate the radiosensitization of ESCC by signal transducer and activator of transcription 3 (STAT3) inhibitor stattic. ECA109, TE13, and KYSE150 cell lines were exposed to hypoxia and treated with stattic or radiation, alone or in combination. Cell proliferation, colony formation, apoptosis, and double-stranded DNA breaks (DSBs) were examined. In addition, ECA109 cells were xenografted into nude mice and treated with radiation and/or stattic. The levels of STAT3, p-STAT3, hypoxia-inducible factor 1α (HIF-1α), and vascular endothelial growth factor (VEGF) in ESCC cells and xenografts were detected by Western blot and immunohistochemical analysis. Our results showed that stattic efficiently radiosensitized ESCC cells and xenografts, especially under hypoxia. Moreover, stattic inhibited STAT3 activation and downregulated HIF-1α and VEGF expression. In conclusion, stattic confers radiosensitivity in ESCC cells in vitro and in vivo and is a potential adjuvant for the radiotherapy of ESCC in the clinical setting.