Search

Hydroquinone (HQ) is one of the major metabolites of benzene and can cause abnormal gene expression. It is a known carcinogen that alters cell cycle disruption and cell proliferation. However, its chemical mechanism remain a mystery. Circular RNAs (circRNAs) are a subtype of noncoding RNAs (ncRNAs) that play a variety of roles in biological processes. Hsa_circ_001944 expression was upregulated in 30 leukemia patients and HQ-induced malignant transformed TK6 cells. Hsa_circ_001944 silencing inhibited the growth of HQ-TK6 cells and halted the cell cycle. The silencing of hsa_circ_0001944 led to increased cell accumulation in G1 versus S phase, increased apoptosis in the sh1944 versus the shNC group, and increased levels of DNA damage (γ-H2AX), leading to cell cycle arrest. In summary, inhibition of hsa_circ_001944 restricted cell growth by inhibiting cell cycle arrest and induced growth of HQ-TK6 cells by modulating PARP1 expression. Hsa_circ_0001944 targeted HuR, which is a kind of RNA-binding protein, to control PARP1 expression via RNAinter, RBPmap, and RBPdb. Fluorescence in situ hybridization combined with immunofluorescent labeling and western blotting experiments showed that hsa_circ_001944 was able to dissociate HuR and PARP1 binding in HQ-TK6 cells, control PARP1 production, and ultimately alter the PARP1/H-Ras pathway.

Limertinib (ASK120067) is a newly developed third-generation EGFR tyrosine kinase inhibitor targeting both sensitizing EGFR and EGFR Thr790Met (T790M) mutations. This study aimed to evaluate the efficacy and safety of limertinib in patients with locally advanced or metastatic EGFR T790M-mutated NSCLC.This is a single-arm, open-label, phase 2b study conducted at 62 hospitals across the People's Republic of China. Patients with locally advanced or metastatic NSCLC with centrally confirmed EGFR T790M mutations in tumor tissue or blood plasma who progressed after first- or second-generation EGFR tyrosine kinase inhibitors or with primary EGFR T790M mutations were enrolled. Patients received limertinib 160 mg orally twice daily until disease progression or unacceptable toxicity. The primary end point was objective response rate (ORR) assessed by independent review committee per the Response Evaluation Criteria in Solid Tumors version 1.1. Secondary end points included disease control rate, progression-free survival (PFS), duration of response (DoR), overall survival, and safety. Safety was assessed according to the National Cancer Institute Common Terminology Criteria for Adverse Events version 4.03.From July 16, 2019, to March 10, 2021, a total of 301 patients were enrolled and started the treatment of limertinib. All patients entered the full analysis set and safety set. By the data cutoff date on September 9, 2021, 76 (25.2%) remained on treatment. The median follow-up time was 10.4 months (range: 0.3-26.3). On the basis of full analysis set, the independent review committee-assessed ORR was 68.8% (95% confidence interval [CI]: 63.2%-74.0%) and disease control rate was 92.4% (95% CI: 88.8%-95.1%). The median PFS was 11.0 months (95% CI: 9.7-12.4), median DoR was 11.1 months (95% CI: 9.6-13.8), and median OS was not reached (95% CI 19.7 months-not evaluable). Objective responses were achieved across all prespecified subgroups. For 99 patients (32.9%) with central nervous system (CNS) metastases, the ORR was 64.6% (95% CI: 54.4%-74.0%), median PFS was 9.7 months (95% CI: 5.9-11.6), and median DoR was 9.6 months (95% CI: 8.1-15.2). For 41 patients who had assessable CNS lesion, the confirmed CNS-ORR was 56.1% (95% CI: 39.7%-71.5%) and median CNS-PFS was 10.6 months (95% CI: 5.6-not evaluable). In safety set, 289 patients (96.0%) experienced at least one treatment-related adverse event (TRAE), with the most common being diarrhea (81.7%), anemia (32.6%), rash (29.9%), and anorexia (28.2%). Grade ≥3 TRAEs occurred in 104 patients (34.6%), with the most common including diarrhea (13.0%), hypokalemia (4.3%), anemia (4.0%), and rash (3.3%). TRAEs leading to dose interruption and dose discontinuation occurred in 24.6% and 2% of patients, respectively. No TRAE leading to death occurred.Limertinib (ASK120067) was found to have promising efficacy and an acceptable safety profile for the treatment of patients with locally advanced or metastatic EGFR T790M-mutated NSCLC.NCT03502850.

Programmed cell death ligand 1 inhibitors combined with chemotherapy has changed the approach to first-line treatment in patients with extensive-stage small cell lung cancer (SCLC). It remained unknown whether adding a programmed cell death 1 (PD-1) inhibitor to chemotherapy provided similar or better benefits in patients with extensive-stage SCLC, which would add evidence on the efficacy of checkpoint inhibitors in the treatment of extensive-stage SCLC.To evaluate the efficacy and adverse event profile of the PD-1 inhibitor serplulimab plus chemotherapy compared with placebo plus chemotherapy as first-line treatment in patients with extensive-stage SCLC.This international, double-blind, phase 3 randomized clinical trial (ASTRUM-005) enrolled patients at 114 hospital sites in 6 countries between September 12, 2019, and April 27, 2021. Of 894 patients who were screened, 585 with extensive-stage SCLC who had not previously received systemic therapy were randomized. Patients were followed up through October 22, 2021.Patients were randomized 2:1 to receive either 4.5 mg/kg of serplulimab (n = 389) or placebo (n = 196) intravenously every 3 weeks. All patients received intravenous carboplatin and etoposide every 3 weeks for up to 12 weeks.The primary outcome was overall survival (prespecified significance threshold at the interim analysis, 2-sided P .012). There were 13 secondary outcomes, including progression-free survival and adverse events.Among the 585 patients who were randomized (mean age, 61.1 [SD, 8.67] years; 104 [17.8%] women), 246 (42.1%) completed the trial and 465 (79.5%) discontinued study treatment. All patients received study treatment and were included in the primary analyses. As of the data cutoff (October 22, 2021) for this interim analysis, the median duration of follow-up was 12.3 months (range, 0.2-24.8 months). The median overall survival was significantly longer in the serplulimab group (15.4 months [95% CI, 13.3 months-not evaluable]) than in the placebo group (10.9 months [95% CI, 10.0-14.3 months]) (hazard ratio, 0.63 [95% CI, 0.49-0.82]; P .001). The median progression-free survival (assessed by an independent radiology review committee) also was longer in the serplulimab group (5.7 months [95% CI, 5.5-6.9 months]) than in the placebo group (4.3 months [95% CI, 4.2-4.5 months]) (hazard ratio, 0.48 [95% CI, 0.38-0.59]). Treatment-related adverse events that were grade 3 or higher occurred in 129 patients (33.2%) in the serplulimab group and in 54 patients (27.6%) in the placebo group.Among patients with previously untreated extensive-stage SCLC, serplulimab plus chemotherapy significantly improved overall survival compared with chemotherapy alone, supporting the use of serplulimab plus chemotherapy as the first-line treatment for this patient population.ClinicalTrials.gov Identifier: NCT04063163.

In cases of limited medical resources, elective total knee arthroplasty (TKA) sometimes needs to be performed after typical work hours. However, surgeon fatigue and logistical factors may potentially affect outcomes. This study aimed to detect whether after-hour procedures impair outcomes after TKA.Elective unilateral TKA from Jan 1, 2016 to Nov 31, 2018 was retrospectively selected and separated into two groups. Procedures started from 8:00 A.M. to 5:29 P.M. were identified as day-time surgeries, whereas those started from 5:30 P.M. to 11:59 P.M. were considered after-hour surgeries. Operative period, Knee Society Score (KSS), range of motion (ROM), total blood loss, length of hospital stay (LOS), and postoperative adverse events and complications were compared. Additionally, the components were evaluated radiologically.A total of 321 patients were selected, including 258 (80.37%) patients in the day-time group and 63 (19.63%) patients in the after-hour group. Operative period, LOS, total blood loss were similar between groups. The overall and each specific incidence of postoperative complications were comparable between the two groups, but the incidence of postoperative vomiting (POV) was higher in the after-hour group. There was no significant difference in knee joint function as shown by the KSS and ROM, both on the 3rd day and at 2 years after surgeries. Radiologically, there were no significant differences between the two groups in the femoral notches (P = 0.592). However, better coronal alignment was detected in the day-time group (P = 0.002), consistent with which there were less outliers (P = 0.033).After-hour TKA procedure does not exert an impact on clinical outcomes, but negatively affects lower limb alignment. Besides, after-hour TKA surgery impairs patients' comfort by increasing POV.

Long-term exposure to benzene or its metabolite, hydroquinone (HQ), can causally contribute to acute myeloid leukemia. Long-noncoding RNAs are essential epigenetic regulators with critical roles in tumor initiation and malignant progression; however, the mechanism by which aberrantly expressed LINC00173 (long intergenic nonprotein coding RNA 173) regulates the pathogenesis of acute myeloid leukemia is not fully understood. Here, we found that the expression of LINC00173 decreased while the expression of DNA methyltransferase 1 (DNMT1) increased, and the methylation of LINC00173 promoter was negatively correlated with LINC00173 expression in GEPIA, CCLE databases, benzene-exposed workers, B-cell non-Hodgkin’s lymphoma, K562, U937, or HQ-induced malignantly transformed TK6 (HQ-MT cells). Furthermore, in 5-aza-2’-deoxycytidine (DNA methyltransferase inhibitor) or trichostatin A (histone deacetylation inhibitor)-treated HQ-MT cells, the expression of LINC00173 was restored by reduced DNA promoter methylation levels. HQ-MT cells with DNMT1 knockout by CRISPR/Cas9 restored the expression of LINC00173 and inhibited the DNA methylation of its promoter as well as enrichment of DNMT1 to promoter. Overexpression of LINC00173 inhibited the expression of DNMT1, cell proliferation, tumor growth, enhanced chemosensitivity to cisplatin, and apoptosis in HQ-MT cells. LINC00173 interacts with DNMT1 to regulate the methylation of LINC00173 promoter. Overall, this study provides evidence that interaction between DNMT1 and LINC00173 regulates the expression of LINC00173 by regulating its promoter methylation level, thus regulating the function of HQ-MT cells in vitro and in vivo, providing a new therapeutic target for benzene-induced tumor.

The upstream regulators of microRNAs were rarely reported. Hydroquinone (HQ) is the main metabolite of benzene, one of the important environmental factors contributing to leukemia and lymphoma. In HQ-induced malignant transformed TK6 (TK6-HT) cells, the expression of PARP-1 and miR-223 were upregulated. When in PARP-1 silencing TK6-HT cells, miR-223 was downregulated and the apoptotic cell number correspondingly increased. In TK6 cells treated with HQ for different terms, the expression of miR-223 and PARP-1 were dynamically observed and found to be decreased and increased, respectively. Trichostatin A could increase the expression of miR-223, then the expression of HDAC1-2 and nuclear factor kappa B were found to be increased, but that of mH2A was decreased. PARP-1 silencing inhibited the protein expression of H3Ac, mH2A, and H3K27ac. By co-immunoprecipitation experiment, PARP-1 and HDAC2 were found to form a regulatory complex. In conclusion, we demonstrated that the upregulation of PARP-1 mediated activation of acetylation to promote the transcription of miR-223 possibly via coregulating with HDAC2, an epigenetic regulation mechanism involved in cell malignant transformation resulting from long-term exposure to HQ, in which course, H3K27ac might be a specific marker for the activation of histone H3, which also gives hints for benzene exposure research.

In the fields of computer vision and graphics, clothing segmentation is a challenging and practical task which is typically implemented in a fine-grained semantic segmentation framework. Unlike the generic semantic segmentation task, clothing segmentation has some domain-specific properties such as diverse appearance variations, non-rigid geometry deformations, and small sample learning. To deal with these points, we propose a semantic locality-preserving segmentation model, which adaptively attaches an original clothing image with a semantically similar (e.g., appearance or pose) auxiliary exemplar by search. Through considering the interactions of the clothing image and its exemplar, more intrinsic knowledge about the locality manifold structures of clothing images is discovered to make the learning process of small sample problem more stable and tractable. Besides, we present a CNN model based on the deformable convolutions to extract the non-rigid geometry-aware features for clothing images. Furthermore, we apply our semantic locality-preserving segmentation model in both image and video cases, resulting in favorable clothing segmentation performance. Experimental results demonstrate the effectiveness of the proposed model against the state-of-the-art approaches.

The high-definition map (HD Map) is one of the essential components for autonomous driving (AV) systems, which is also helpful to realize the intelligent transportation system. For autonomous vehicles, the HD Map is closely connected with the modules of high-precision localization, environment perception, and decision-making control in the AV systems. Although many mapping companies have launched HD Map products, such as HERE, Tomtom, Civil Maps, Baidu and NavInfo, there is currently no globally unified standard for HD Map. And these map companies are almost aimed at large areas such as regions and cities, and few of them are aimed at specific scenes. In addition, the mapping and maintenance costs of these HD Maps are also very expensive. In this paper, we develop a HD Map in a specific scene, which can not only better serve the AV application in the scene, but also reduce the cost of mapping, maintenance and updating. First, we give the definition classification of specific scenes. Next, we put forward a novel multi-client and multi-layer HD Map architecture for specific scenes. Finally, we introduce the key technologies for mapping of each layer, and verify the accuracy of the positioning layer by comparative experiment of relative accuracy.

Model performance can be further improved with the extra guidance apart from the one-hot ground truth. To achieve it, recently proposed recollection-based methods utilize the valuable information contained in the past training history and derive a ``recollection'' from it to provide data-driven prior to guide the training. In this article, we focus on two fundamental aspects of this method, i.e., recollection construction and recollection utilization. Specifically, to meet the various demands of models with different capacities and at different training periods, we propose to construct a set of recollections with diverse distributions from the same training history. After that, all the recollections collaborate together to provide guidance, which is adaptive to different model capacities, as well as different training periods, according to our similarity-based elastic knowledge distillation (KD) algorithm. Without any external prior to guide the training, our method achieves a significant performance gain and outperforms the methods of the same category, even as well as KD with well-trained teacher. Extensive experiments and further analysis are conducted to demonstrate the effectiveness of our method.

Background. Accurate interpretation of lung function tests requires appropriate spirometry reference values derived from large-scale population-specific epidemiological surveys. The aim of this cross-sectional study was to establish normal spirometric values for the population of healthy, nonsmoking Han Chinese adults residing in Zhejiang province, China. Methods. We measured lung function parameters such as forced expiratory volume in 1 s, forced vital capacity, peak expiratory flow, maximal midexpiratory flow, and diffusion capacity for carbon monoxide and considered age, height, and weight as independent factors that may modify these parameters. The clinical data were divided into the study arm and validation group. The study arms were used to construct predictive equations using stepwise multiple linear regression, and data from the validation group were used to assess the robustness of the equations. Results. The 3866 participants were randomized into a study arm (n = 1,949) and a validation arm (n = 1,917). Lung function parameters had a negative association with age and a positive association with height. Data from the two groups were similar. Predictive equations were constructed from the study arm, and the validation group was used to test the feasibility of the reference equations. Conclusions. The reference values we derived can be used to evaluate lung function in this cohort in both epidemiological studies and clinical practice.

Photoelectrocatalytic methanol oxidation to formaldehyde by using electric/solar energy is a promising technology. In this work, Ag nanoparticles were decorated on Mo-doped BiVO4 to synthesize Ag/Mo: BiVO4 catalyst. As expected, benefitting from Mo-dopant and plasmonic effect of Ag nanoparticles, formaldehyde production of Ag/Mo: BiVO4 photoanode was 2.2 times higher than pristine BiVO4 photoanode. Faradaic efficiency increased from 19% to 49% in comparison with the pure BiVO4 photoanode.

Exposure to benzene or its metabolite hydroquinone (HQ) is a risk factor for a series of myeloid malignancies, and long noncoding RNAs play an important role in the process of pathogenesis. Urothelial cancer-associated 1 (UCA1) functions as an oncogene in the development of acute myeloid leukemia. However, the association between DNMT1 and UCA1 with benzene or HQ exposure has not been explored. We characterized UCA1 expression in cells briefly exposed to HQ (HQ-ST cells) and HQ-induced malignantly transformed (TK6-HT cells) treated with 5-aza-2'-deoxycytidine (5-AzaC) or trichostatin A (TSA). Compared to that in control cells, UCA1 expression was increased, whereas DNMT1 was decreased in HQ-ST cells and TK6-HT cells treated with 5-AzaC or TSA. Moreover, UCA1 expression was also upregulated and positively correlated with benzene exposure time in benzene-exposed workers. Furthermore, the expression of UCA1 was negatively associated with the DNA methylation level of its promoter in benzene-exposed workers. DNMT1 rather than DNMT3b knockout in TK6-HT cells activated the expression of UCA1 by inducing its promoter hypomethylation. These results suggest that benzene or HQ exposure leads to UCA1 upregulation via DNA hypomethylation in the UCA1 promoter, which is mediated by DNMT1.

Accurate detection of objects in perception system is a basic task for autonomous vehicles to operate reliably. The current mainstream paradigms for perception and prediction are able to decompose into three sequential subtasks: detection, tracking and prediction. Much of the previous works focus on one of these subtasks and internal relationship between the various subtasks are therefore ignored. In this work we propose an end-to-end model that is able to jointly carry out 3D detection and motion prediction in the context of autonomous vehicles, which can benefit from joint optimization of two tasks. Considering computational consumption and memory cost, our approach utilizes 2D convolutions across space over a bird's eye view (BEV) representation of the 3D world, not 3D convolutions which require more inference time for a single point cloud. In this paper, the key model is the sequential multi-frame point clouds channels based on BEV, which can help perform object detection and learn temporal information at the same time. Experiments on nuScenes dataset show that the proposed network achieves state-of-the-art results, and the trade-off between accuracy and efficiency of two tasks are obtained. Additionally, we validate the effectiveness of our models through the ablation studies. Specifically, by sharing network, parameters and detection results we can perform two tasks within 32 FPS.

Solving the optimization problem to approach a Nash Equilibrium point plays an important role in imperfect information games, e.g., StarCraft and poker. Neural Fictitious Self-Play (NFSP) is an effective algorithm that learns approximate Nash Equilibrium of imperfect-information games from purely self-play without prior domain knowledge. However, it needs to train a neural network in an off-policy manner to approximate the action values. For games with large search spaces, the training may suffer from unnecessary exploration and sometimes fails to converge. In this paper, we propose a new Neural Fictitious Self-Play algorithm that combines Monte Carlo tree search with NFSP, called MC-NFSP, to improve the performance in real-time zero-sum imperfect-information games. With experiments and empirical analysis, we demonstrate that the proposed MC-NFSP algorithm can approximate Nash Equilibrium in games with large-scale search depth while the NFSP can not. Furthermore, we develop an Asynchronous Neural Fictitious Self-Play framework (ANFSP). It uses asynchronous and parallel architecture to collect game experience and improve both the training efficiency and policy quality. The experiments with th e games with hidden state information (Texas Hold’em), and the FPS (firstperson shooter) games demonstrate effectiveness of our algorithms.

Cryptococcemia is a life-threatening fungal infection. Sometimes, it is hard to diagnose. The studies to describe the characteristics of cryptococcemia specifically were limited. We performed this retrospective analysis in a Chinese hospital during 2002–2015, including 85 cryptococcemia cases and 52 Cryptococcus spp. isolates. The species, mating type, antifungal susceptibility and multilocus sequence typing of Cryptococcus spp. were determined. C. neoformans var. grubii MATα of sequence type (ST) 5 is the representative strain of cryptococcemia, accounting for 51 isolates. The MIC50/90 values were 0.5/0.5, 1.0/1.0, 2.0/4.0, ≤0.06/0.25, and ≤0.06/≤0.06 μg/ml for amphotericin B, flucytosine, fluconazole, itraconazole, and voriconazole, respectively. Cryptococcemia was the first diagnostic proof of cryptococcosis in 37 patients (43.5%, 37/85). Compared with the patients initially diagnosed of cryptococcosis in other sites (mainly cerebrospinal fluid), the patients firstly diagnosed by blood culture had prolonged time from admission to diagnosis of cryptococcosis (9 days vs. 2 days, P < .001) and higher 30-day mortality (54.1% vs. 20.8%, P = .003), while fewer symptoms of meningitis (45.9% vs. 100%, P < .001). For the patients receiving lumbar puncture, the occurrence of meningitis was similar between the patients firstly diagnosed by blood culture and those firstly diagnosed in other sites (94.1% vs. 100%, P = .26). However, the patients first diagnosed by blood culture had lower baseline intracranial pressure (250 mm H2O vs. 342.5 mm H2O, P = .001). In conclusion, patients with cryptococcemia as the first diagnostic proof of cryptococcosis usually had neglected subtle symptoms of meningitis, which may result in delayed diagnosis and catastrophic outcome.

Hydroquinone (HQ), one of the major metabolites of benzene, can induce aberrant gene expression. MiR-155, a tumor activator, participates in various biological processes, including DNA damage response. However, the molecular mechanism of aberrant miR-155 expression is still not completely elucidated. Here, we investigated the mechanism of abnormal expression of miR-155 induced by poly(ADP-ribose)polymerase-1 (PARP-1) expression in HQ-treated TK6 lymphoblastoid cells. We examined the expression of genes related to abnormal expression of miR-155 to explore the reason for this phenomenon. The results of the present study showed that miR-155 was significantly increased and reactive oxygen species (ROS) were decreased in cells treated with HQ for 72 h compared with PBS-treated cells. Meanwhile, E4F1, PARP-1 and PARP-1 related co-regulators (NF-κB, HDAC1, and HDAC2), acetylated histone H3 (H3Ac) were increased in a concentration-dependent manner. Experiments for treatment with 5-AzaC (DNMTs inhibitor), TSA (HDACs inhibitor), DOX (to activate PARP-1) or MG132 (proteasome inhibitor) revealed that the MBDs and PARP-1 was positively associated with miR-155 expression. Moreover, in cells treated with HQ in conjunction with PARP-1 knockdown, expression of miR-155, H3Ac and MBD2 protein were decreased, compared with negative control. In conclusion, PARP-1 activates expression of miR-155 via acetylation by regulating MBD2 in TK6 cells exposed to HQ.

9106 Background: ASK120067 (Limertinib) is a newly developed third-generation epidermal growth factor receptor (EGFR) tyrosine kinase inhibitor (TKI) targeting both sensitizing EGFR and EGFR T790M mutations. This study aimed to evaluate the efficacy and safety of ASK120067 in patients with locally advanced or metastatic EGFR T790M mutated non-small cell lung cancer (NSCLC). Methods: This study was a single-arm, open-label, phase 2b study conducted at 62 hospitals across China. Patients with locally advanced or metastatic NSCLC with centrally confirmed EGFR T790M mutations in tumor tissue or blood plasma who progressed after first or second generation EGFR TKIs or with primary EGFR T790M mutations were enrolled. Patients received ASK120067 160mg orally twice daily, until disease progression, or unacceptable toxicity. The primary endpoint was objective response rate (ORR) assessed by Independent Review Committee (IRC) per RECIST1.1. Secondary endpoints included disease control rate (DCR), progression-free survival (PFS), duration of response (DOR), overall survival (OS), and safety. Safety was assessed according to CTCAE 4.03. Results: Between June 24th, 2019 and Feb 25th, 2021 301 patients were enrolled and received ASK120067 treatment. All patients entered the full analysis set (FAS) and safety set (SS). A total of 99 (32.9%) patients had central nervous system (CNS) metastases at baseline. By the data cutoff date on Sep 9th, 2021, 76 (25.2%) remained on treatment. The median follow-up time was10.4 (range 0.3-26.3) months. Based on FAS, the IRC-assessed ORR was 68.8% (95%CI 63.2%-74.0%) and DCR was 92.4% (95%CI 88.8%-95.1%). The median PFS was 11.0 (95%CI 9.7-12.4) months, median DOR was 11.1 (95%CI 9.6-13.8) months, and median OS was not reached (NR) (95%CI 19.7 months-NR). Objective responses were achieved across all pre-specified subgroups. For 99 patients with CNS metastases, the ORR was 64.6% (95%CI 54.4%-74.0%), median PFS was 9.7 (95%CI 5.9-11.6) months, and median DOR was 9.6 (95%CI 8.1-15.2) months. For 41 patients who had evaluable CNS lesion, the confirmed CNS-ORR was 56.1% (95%CI, 39.7%-71.5%) and median CNS-PFS was 10.6 (95%CI 5.6-NE) months. In SS, 289 (96.0%) patients experienced at least one adverse drug reaction (ADR), with the most common being diarrhea (81.7%), anemia (32.6%), rash (29.9%) and appetite decrease (28.2%). Grade ≥3 ADRs occurred in 104 (34.6%) patients, and the most common included diarrhea (13.0%), hypokalemia (4.3%), anemia (4.0%) and rash (3.3%). ADRs leading to dose interruption and dose discontinuation occurred 24.6% and 2% of patients, respectively. No ADR leading to death occurred. Conclusions: ASK120067 demonstrated promising efficacy and an acceptable safety profile for the treatment of patients with locally advanced or metastatic EGFR T790M mutated NSCLC. Clinical trial information: NCT03502850.

Different from many other attributes, facial expression can change in a continuous way, and therefore, a slight semantic change of input should also lead to the output fluctuation limited in a small scale. This consistency is important. However, current Facial Expression Recognition (FER) datasets may have the extreme imbalance problem, as well as the lack of data and the excessive amounts of noise, hindering this consistency and leading to a performance decreasing when testing. In this paper, we not only consider the prediction accuracy on sample points, but also take the neighborhood smoothness of them into consideration, focusing on the stability of the output with respect to slight semantic perturbations of the input. A novel method is proposed to formulate semantic perturbation and select unreliable samples during training, reducing the bad effect of them. Experiments show the effectiveness of the proposed method and state-of-the-art results are reported, getting closer to an upper limit than the state-of-the-art methods by a factor of 30\% in AffectNet, the largest in-the-wild FER database by now.

The multi-clearances of breaking type steering trapezoidal mechanism joints influences vehicle steering stability. Hence, to ascertain the influence of clearance value on steering stability, this paper takes the steering mechanism of a certain vehicle type as a prototype that can be simplified into a planar six-bar linkage, then establishes the system dynamic differential equations after considering the two clearances of tie rods and the steering knuckle arms. The influence of the clearance parameters on the movement stability of the steering mechanism is studied using a numerical computation method. Results show that when the two clearances are equal, the planar movement of the tie rods changes from period-doubling to chaos as the clearances increase. When the two clearances are 0.25 mm and 1.5 mm respectively, the planar movements of the two side tie rods come into chaos, causing the steering stability to deteriorate. Moreover, with the increase of clearances, turning moment fluctuates more intensively and the peak value increases.

Benzene exposure is a risk factor of acute myeloid leukemia (AML), during such carcinogenesis long non-coding RNAs (lncRNAs) are important epigenetic regulators. HOTAIRM1 (HOXA transcript antisense RNA, myeloid-specific 1) plays an indispensable role in the development of AML. Hydroquinone (HQ) is one major metabolite of benzene and its ideal replacement in toxicology research. But the influence of benzene or HQ on HOTAIRM1 expression in AML associated pathway is still unclear. In the TK6 cells with short-term exposure to HQ (HQ-ST cells) or long term HQ exposure induced malignant transformed TK6 cells (HQ-MT cells), the relationship between DNMT3b and HOTAIRM1 was explored. Comparing to counterparts, HOTAIRM1 expression was increased firstly and then decreased in HQ-ST cells, and definitely decreased in HQ-MT cells; while the expression change tendency of DNMT3b was in contrast to that of HOTAIRM1. Moreover, the average HOTAIRM1 expression of 17 paired workers being exposed to benzene within 1.5 years was increased, but that of the remaining 92 paired workers with longer exposure time was decreased. Furthermore, in 5-AzaC (DNA methyltransferase inhibitor) or TSA (histone deacetylation inhibitor) treated HQ-MT cells, the expression of HOTAIRM1 was restored by reduced DNA promoter methylation levels. HQ-MT cells with DNMT3b knockout by CRISPR/Cas9 displayed the promoter hypomethylation and the increase of HOTAIRM1, also confirmed in benzene exposure workers. These suggest that long term exposure to HQ or benzene might induce the increase of DNMT3b expression and the promoter hypermethylation to silence the expression of HOTAIRM1, a possible tumor-suppressor in the AML associated carcinogenesis pathway.

Aiming at the problems of little expansion and small bending angle of the soft robot, a modular soft robot two-stage driven mechanism is designed. The mechanism including expansion module and flexural module, expansion module expansion rate can reach 50%, and can provide greater driving force, bending module bending angle can reach 70°.The mechanism can realize three degrees of freedom motion of space by expansion module and flexural module. The mechanism is modular, so it has high controllability and can connect multiple modules in series. By means of screw theory, the kinematic equation of the mechanism is derived. According to the material phase change characteristic of SMA spring and the thermodynamic theories, the relationship between the current and the deformation of SMA springs can be obtained. The mechanism is analysed by simulation software. It is concluded that the design of the mechanism solves the problems of modularization, less expansion and small bending angle of the soft robot, and the feasibility of the mechanism model is verified.

Patients with rheumatoid arthritis (RA) taking long-term immunosuppressive drugs are more susceptible to opportunistic infections, such as cryptococcosis. A 65-year-old woman was transferred to our hospital for rapidly progressing pulmonary lesions identified by lung computed tomography. She had a 7-year history of RA and had been prescribed methotrexate and glucocorticoids for 10 months. Additionally, our patient had a history of environmental exposure to house renovation lasting approximately 1 week before onset. Her serological test results and histopathological examination confirmed the diagnosis of pulmonary cryptococcosis (PC). The patient recovered well after 6 months of fluconazole treatment. In addition, we summarized 28 reported cases of RA patients with PC and found that older age might be a risk factor for cryptococcal infection in RA patients. The most common location for pulmonary lesions was the lower lobe, and the most common radiologic manifestations were nodules. Detection of cryptococcal capsular polysaccharide antigen was important for diagnosis. Patients undergoing antirheumatic therapy should avoid exposure to Cryptococcus.

Hydroquinone (HQ), one of the main metabolites of benzene, is a well-known human leukemogen. However, the specific mechanism of how benzene or HQ contributes to the development of leukemia is unknown. In a previous study, we demonstrated the upregulation of DNA methyltransferase (DNMT) expression in HQ-induced malignant transformed TK6 (HQ-TK6) cells. Here, we investigated whether a regulatory loop between the long noncoding RNA FAS-AS1 and DNMT3b exists in HQ-TK6 cells and benzene-exposed workers. We found that the expression of FAS-AS1 was downregulated in HQ-TK6 cells and workers exposed to benzene longer than 1.5 years via histone acetylation, and FAS-AS1 expression was negatively correlated with the time of benzene exposure. Restoration of FAS-AS1 in HQ-TK6 cells promoted apoptosis and inhibited tumorigenicity in female nude mice. Interestingly, treatment with a DNMT inhibitor (5-aza-2-deoxycytidine), histone deacetylase inhibitor (trichostatin A), or DNMT3b knockout led to increased FAS-AS1 through increased H3K27ac protein expression in HQ-TK6 cells, and DNMT3b knockout decreased H3K27ac and DNMT3b enrichment to the FAS-AS1 promoter region, which suggested that DNMT3b and/or histone acetylation involve FAS-AS1 expression. Importantly, restoration of FAS-AS1 resulted in reduced expression of DNMT3b and SIRT1 and increased expression of FAS in both HQ-TK6 cells and xenograft tissues. Moreover, the average DNMT3b expression in 17 paired workers exposed to benzene within 1.5 years was decreased, but that of the remaining 103 paired workers with longer exposure times was increased. Conversely, DNMT3b was negatively correlated with FAS-AS1 expression. Both FAS-AS1 and DNMT3b influenced the enrichment of H3K27ac in the FAS promoter region by regulating the expression of SIRT1, consequently upregulating FAS expression. Taken together, these observations demonstrate crosstalk between FAS-AS1 and DNMT3b via a mutual inhibition loop and indicate a new mechanism by which FAS-AS1 regulates the expression of FAS in benzene-related carcinogenesis.

Urban traffic systems worldwide are suffering from severe traffic safety problems. Traffic safety is affected by many complex factors, and heavily related to all drivers' behaviors involved in traffic system. Drivers with aggressive driving behaviors increase the risk of traffic accidents. In order to manage the safety level of traffic system, we propose Driving Safety Credit inspired by credit score in financial security field, and design a scoring method using driver's trajectory data and violation records. First, we extract driving habits, aggressive driving behaviors and traffic violation behaviors from driver's trajectories and traffic violation records. Next, we train a classification model to filtered out irrelevant features. And at last, we score each driver with selected features. We verify our proposed scoring method using 40 days of traffic simulation, and proves the effectiveness of our scoring method.