Search

Supplementary Figure 1. Brief schematic of the strategy to identify salivary miRNA biomarkers to detect resectable pancreatic cancer.

Early surgery is vital in the treatment of pancreatic cancer, which is often fatal. However, there is currently no useful noninvasive biomarker to screen for pancreatic cancer. Studies have documented that many salivary molecules can be used to detect systemic diseases. We investigated whether salivary miRNAs are useful biomarkers for detecting resectable pancreatic cancer. Using an Agilent microarray, salivary miRNAs were profiled from saliva samples of 8 patients with resectable pancreatic cancer and 8 healthy controls. Candidate biomarkers identified in the profiles were subjected to validation using quantitative PCR and an independent sample set of 40 patients with pancreatic cancer, 20 with benign pancreatic tumors (BPT), and 40 healthy controls. The validated salivary miRNA biomarkers were evaluated within three discriminatory categories: pancreatic cancer versus healthy control, pancreatic cancer versus BPT, and pancreatic cancer versus noncancer (healthy control + BPT). miR-3679-5p showed significant downregulation in the pancreatic cancer group within the three categories (P = 0.008, 0.007, and 0.002, respectively), whereas miR-940 showed significant upregulation in pancreatic cancer (P = 0.006, 0.004, and 0.0001, respectively). Logistic regression models combining the two salivary miRNAs were able to distinguish resectable pancreatic cancer within the three categories, showing sensitivities of 72.5%, 62.5%, and 70.0% and specificities of 70.0%, 80.0%, and 70.0%, respectively. Salivary miR-3679-5p and miR-940 possess good discriminatory power to detect resectable pancreatic cancer, with reasonable specificity and sensitivity. This report provides a new method for the early detection of pancreatic cancer and other systemic diseases by assessing salivary miRNAs. Cancer Prev Res; 8(2); 165–73. ©2014 AACR.

Supplementary material 4. Normalised microarray data in the discovery phase.

Supplementary material 3. Detailed demographic, clinical and experimental information of the enrolled participants including patients with resectable pancreatic cancer and benign pancreatic tumor, and healthy controls.

Supplementary material 2. Pancreatic cancer staging based on The American Joint Committee on Cancer: 7th Edition on Cancer Staging.

Supplementary table 1. Demographic and clinical information of enrolled participants.

Supplementary material 1. The definition of pancreatic cancers deemed resectable or borderline resectable according to the National Comprehensive Cancer Network: Practice Guidelines in Oncology (v. 2.2010).

Normal erythropoiesis requires the precise regulation of gene expression patterns, and transcription cofactors play a vital role in this process. Deregulation of cofactors has emerged as a key mechanism contributing to erythroid disorders. Through gene expression profiling, we found HES6 as an abundant cofactor expressed at gene level during human erythropoiesis. HES6 physically interacted with GATA1 and influenced the interaction of GATA1 with FOG1. Knockdown of HES6 impaired human erythropoiesis by decreasing GATA1 expression. Chromatin immunoprecipitation and RNA sequencing revealed a rich set of HES6- and GATA1-co-regulated genes involved in erythroid-related pathways. We also discovered a positive feedback loop composed of HES6, GATA1 and STAT1 in the regulation of erythropoiesis. Notably, erythropoietin (EPO) stimulation led to up-regulation of these loop components. Increased expression levels of loop components were observed in CD34+ cells of polycythemia vera patients. Interference by either HES6 knockdown or inhibition of STAT1 activity suppressed proliferation of erythroid cells with the JAK2V617F mutation. We further explored the impact of HES6 on polycythemia vera phenotypes in mice. The identification of the HES6–GATA1 regulatory loop and its regulation by EPO provides novel insights into human erythropoiesis regulated by EPO/EPOR and a potential therapeutic target for the management of polycythemia vera.

As an effective soil amendment, biochars require a comprehensive ecological evaluation before they can be widely used in agriculture because endogenous contaminants, such as environmentally persistent free radicals (EPFRs), certainly pose an ecological risk to soil invertebrates. In this study, Caenorhabditis elegans (C. elegans) was used as a model organism to investigate the neurotoxicity of two rice straw biochars pyrolyzed at 500 and 700 °C. After 24 h exposure to unwashed biochar, washed biochar, and leaching fluids (supernatants), the neurobehavioral parameters of C. elegans were determined in a liquid toxicity test. The results showed that the washed 700 °C biochar particles significantly impaired locomotion and prolonged the defecation interval at a biochar concentration of 4 g·well−1, while the unwashed biochar and supernatants caused no apparent impairment. Supporting this, electron paramagnetic resonance (EPR) results showed that the intensity of EPFRs in unwashed 700 °C biochar was stronger than that of the corresponding washed particles. This indicates that, in the liquid test, the EPR signal alone is not indicative of particle toxicity. The accessibility and activity of the EPFRs should be considered. Dissolved organic matter (DOM) was observed in the leaching fluids. The neurotoxic activity of the washed biochar was alleviated after the re-addition of leaching fluids to the washed biochar, suggesting that the dissolved organic materials modulate the reactivity of the EPFRs in the liquid phase. This study suggests that the leaching process may increase the risk of biochar when used in the field environment. National Natural Scientific Foundation of China Yunnan Province Basic Research Project NSFC-NCN

Background Transient receptor potential vanilloid subtype 1 (TRPV1) plays a pivotal role in neurons and is closely related to pain transduction. However, the role of TRPV1 in pan-cancer remains unclear. Methods Data visualization was performed using TCGA, GETx, GEPIA2, TIMER, TISIDB, DiseaseMeth, GeneMANIA, GSEA, and Depmap Portal database and R language. Results We explored the pan-cancer expression patterns and prognostic value of TRPV1 across multiple databases and found that TRPV1 served as a tumor suppressor in most cancers. In particular, genetic alteration and DNA methylation analysis across 33 tumors revealed that gene deletion and DNA hypermethylation may contribute to its downregulation. We further constructed a transcription factors-TRPV1 regulatory network and revealed a series of key upstream transcription factors of TRPV1 in certain cancers. Additionally, we studied the relationship between TRPV1 levels and the tumor microenvironment, immune cells, immune checkpoints, and its sensitivity to small-molecule inhibitors. Conclusions Our study curated both the genetic and epigenetic status of TRPV1 and its regulatory networks in cancers and highlighted that TRPV1 could serve as a prognostic biomarker and is correlated with immunotherapy and chemotherapy.

As yet, there is no unified method of treatment for the evaluation and management of gastric low-grade intraepithelial neoplasia (LGIN) worldwide.Patients with gastric LGIN who had been treated withA total of 309 patients with LGIN were randomly divided into the training groups and the validation groups. LASSO regression analysis and multivariable logistic regression identified that 6 variables including gender, size, location, borderline, number, and erosion were independent risk factors. The nomogram model displayed good discrimination with a C-index of .765 (95% confidence interval: .702-.828). The accuracy and reliability of the model were also verified by an AUC of .764 in the training group and .757 in the validation group. Meanwhile, the calibration curve and the DCA suggested that the predictive nomogram had promising accuracy and clinical utility.A predictive nomogram model was constructed and proved to be clinically applicable to identify high-risk groups with possible pathologic upgrade in patients with gastric LGIN. Since it is regarded that strengthening follow-up or endoscopic treatment of high-risk patients may contribute to improving the detection rate or reducing the incidence of gastric cancer, the predictive nomogram model provides a reliable basis for the treatment of LGIN.

Analogical reasoning is the process of discovering and mapping correspondences from a target subject to a base subject. As the most well-known computational method of analogical reasoning, Structure-Mapping Theory (SMT) abstracts both target and base subjects into relational graphs and forms the cognitive process of analogical reasoning by finding a corresponding subgraph (i.e., correspondence) in the target graph that is aligned with the base graph. However, incorporating deep learning for SMT is still under-explored due to several obstacles: 1) the combinatorial complexity of searching for the correspondence in the target graph; 2) the correspondence mining is restricted by various cognitive theory-driven constraints. To address both challenges, we propose a novel framework for Analogical Reasoning (DeepGAR) that identifies the correspondence between source and target domains by assuring cognitive theory-driven constraints. Specifically, we design a geometric constraint embedding space to induce subgraph relation from node embeddings for efficient subgraph search. Furthermore, we develop novel learning and optimization strategies that could end-to-end identify correspondences that are strictly consistent with constraints driven by the cognitive theory. Extensive experiments are conducted on synthetic and real-world datasets to demonstrate the effectiveness of the proposed DeepGAR over existing methods., 22nd IEEE International Conference on Data Mining (ICDM 2022)

As a type of effective soil amendment, biochars require comprehensive ecological assessment before their widespread agricultural applications, since the endogenous contaminants, especially environmentally persistent free radicals (EPFRs) in biochars certainly pose an ecological risk to soil invertebrates. In this study, Caenorhabditis elegans (C. elegans) was chosen as a model organism to assess the neurotoxicity of three rice straw biochars pyrolyzed at 500, 700, and 1000°C. After 24 h-exposure to unwashed biochars, washed biochars, and leaching liquids (supernatants), neurobehavioral parameters of L4 C. elegans larvae were determined in a liquid toxicity test. The results showed that the washed 700°C biochar particles significantly impaired the locomotion and prolonged the defecation interval at biochar concentration of 4 g·well− 1, whereas unwashed biochars and supernatants did not cause obvious impairment. At the same time, the electron paramagnetic resonance (EPR) results showed that the intensity of EPFRs in unwashed 700°C biochar was stronger than that of corresponding washed particles. It suggests that, in the liquid test, EPR signal alone does not indicate the toxicity of the particle. The accessibility and activity of EPFRs should be considered. Dissolved organic matter (DOM) has great significance in the fate, transport, and toxicity of environmental pollutants, which is beneficial for C. elegans to live longer and increase their reproductive ability. In this paper, the neurobehaviors of washed biochar were alleviated after adding supernatant back, which indicated that DOM could modulate the reactivity of EPFRs in the liquid phase.

Background: At present, there is no unified treatment for the evaluation and management of gastric low-grade intraepithelial neoplasia (LGIN) all over the world.Methods: Patients who were Helicobacter pylori eradicated, with low-grade gastric intraepithelial neoplasia were gathered.Several demographic and clinicopathological characteristics were described and analyzed retrospectively by LASSO regression analysis and multivariable logistic regression. Then the predictive nomogram was established. C-index, the area under the receiver operating characteristic curve (AUC) , calibration plot and decision curve analysis (DCA) were used to evaluate the accuracy and reliability of the model. Results: A total of 309 patients with LGIN were included, divided into training groups and validation groups randomly. LASSO regression analysis and multivariable logistic regression showed that six variables, gender, size, location, border line, number and erosion were independent risk factors for progression of gastric LGIN. The nomogram model displayed good discrimination with a C-index of 0.765 (95% confidence interval: 0.702–0.828). High C-index value of 0.768 could still be reached in the internal validation. The accuracy and reliability of the model was also verified by the AUC of 0.764 in the training group and 0.757 in the validation group. The calibration curve showed the model was in good agreement with the actual results as well. Decision curve analysis suggested that the predictive nomogram had clinical utility. Conclusions: A predictive nomogram model was successfully established and proved to identify high-risk groups with possible pathologic upgrade in patients with gastric LGIN. It suggested that after identifying high-risk patients, strengthening follow-up or endoscopic treatment may benefit in improving the detection rate or reducing the incidence of gastric cancer, which providing a reliable basis for the treatment of LGIN.

Fluxapyroxad (Figure 1), a broad-spectrum pyrazole-carboxamide fungicide, is commonly employed for the treatment of a large variety of commercial crops. More specifically, the application of Fluxap...

The success of training accurate models strongly depends on the availability of a sufficient collection of precisely labeled data. However, real-world datasets contain erroneously labeled data samples that substantially hinder the performance of machine learning models. Meanwhile, well-labeled data is usually expensive to obtain and only a limited amount is available for training. In this paper, we consider the problem of training a robust model by using large-scale noisy data in conjunction with a small set of clean data. To leverage the information contained via the clean labels, we propose a novel self-paced robust learning algorithm (SPRL) that trains the model in a process from more reliable (clean) data instances to less reliable (noisy) ones under the supervision of well-labeled data. The self-paced learning process hedges the risk of selecting corrupted data into the training set. Moreover, theoretical analyses on the convergence of the proposed algorithm are provided under mild assumptions. Extensive experiments on synthetic and real-world datasets demonstrate that our proposed approach can achieve a considerable improvement in effectiveness and robustness to existing methods.

With the advance of sensor technologies, the Multivariate Time Series classification (MTSC) problem, perhaps one of the most essential problems in the time series data mining domain, has continuously received a significant amount of attention in recent decades. Traditional time series classification approaches based on Bag-of-Patterns or Time Series Shapelet have difficulty dealing with the huge amounts of feature candidates generated in high-dimensional multivariate data but have promising performance even when the training set is small. In contrast, deep learning based methods can learn low-dimensional features efficiently but suffer from a shortage of labelled data. In this paper, we propose a novel MTSC model with an attentional prototype network to take the strengths of both traditional and deep learning based approaches. Specifically, we design a random group permutation method combined with multi-layer convolutional networks to learn the low-dimensional features from multivariate time series data. To handle the issue of limited training labels, we propose a novel attentional prototype network to train the feature representation based on their distance to class prototypes with inadequate data labels. In addition, we extend our model into its semi-supervised setting by utilizing the unlabeled data. Extensive experiments on 18 datasets in a public UEA Multivariate time series archive with eight state-of-the-art baseline methods exhibit the effectiveness of the proposed model.

Sound Event Early Detection (SEED) is an essential task in recognizing the acoustic environments and soundscapes. However, most of the existing methods focus on the offline sound event detection, which suffers from the over-confidence issue of early-stage event detection and usually yield unreliable results. To solve the problem, we propose a novel Polyphonic Evidential Neural Network (PENet) to model the evidential uncertainty of the class probability with Beta distribution. Specifically, we use a Beta distribution to model the distribution of class probabilities, and the evidential uncertainty enriches uncertainty representation with evidence information, which plays a central role in reliable prediction. To further improve the event detection performance, we design the backtrack inference method that utilizes both the forward and backward audio features of an ongoing event. Experiments on the DESED database show that the proposed method can simultaneously improve 13.0\% and 3.8\% in time delay and detection F1 score compared to the state-of-the-art methods., Comment: ICASSP 2022

In many high-stakes applications of machine learning models, outputting only predictions or providing statistical confidence is usually insufficient to gain trust from end users, who often prefer a transparent reasoning paradigm. Despite the recent encouraging developments on deep networks for sequential data modeling, due to the highly recursive functions, the underlying rationales of their predictions are difficult to explain. Thus, in this paper, we aim to develop a sequence modeling approach that explains its own predictions by breaking input sequences down into evidencing segments (i.e., sub-sequences) in its reasoning. To this end, we build our model upon convolutional neural networks, which, in their vanilla forms, associates local receptive fields with outputs in an obscure manner. To unveil it, we resort to case-based reasoning, and design prototype modules whose units (i.e., prototypes) resemble exemplar segments in the problem domain. Each prediction is obtained by combining the comparisons between the prototypes and the segments of an input. To enhance interpretability, we propose a training objective that delicately adapts the distribution of prototypes to the data distribution in latent spaces, and design an algorithm to map prototypes to human-understandable segments. Through extensive experiments in a variety of domains, we demonstrate that our model can achieve high interpretability generally, together with a competitive accuracy to the state-of-the-art approaches.

Recent multilingual pre-trained language models have achieved remarkable zero-shot performance, where the model is only finetuned on one source language and directly evaluated on target languages. In this work, we propose a self-learning framework that further utilizes unlabeled data of target languages, combined with uncertainty estimation in the process to select high-quality silver labels. Three different uncertainties are adapted and analyzed specifically for the cross lingual transfer: Language Heteroscedastic/Homoscedastic Uncertainty (LEU/LOU), Evidential Uncertainty (EVI). We evaluate our framework with uncertainties on two cross-lingual tasks including Named Entity Recognition (NER) and Natural Language Inference (NLI) covering 40 languages in total, which outperforms the baselines significantly by 10 F1 on average for NER and 2.5 accuracy score for NLI., Accepted to EMNLP 2021

Hydrogen is a renewable energy source with various features, clean, carbon-free, high energy density, which is being recognized internationally as a “future energy.” The US, the EU, Japan, South Korea, China, and other countries or regions are gradually clarifying the development position of hydrogen. The rapid development of the hydrogen energy industry requires more hydrogenation infrastructure to meet the hydrogenation need of hydrogen fuel cell vehicles. Nevertheless, due to the frequent occurrence of hydrogen infrastructure accidents, their safety has become an obstacle to large-scale construction. This paper analyzed five sizes (diameters of 0.068 mm, 0.215 mm, 0.68 mm, 2.15 mm, and 6.8 mm) of hydrogen leakage in the hydrogen fueling station using Quantitative Risk Assessment (QRA) and HyRAM software. The results show that unignited leaks occur most frequently; leaks caused by flanges, valves, instruments, compressors, and filters occur more frequently; and the risk indicator of thermal radiation accident and structure collapse accident caused by overpressure exceeds the Chinese individual acceptable risk standard and the risk indicator of a thermal radiation accident and head impact accident caused by overpressure is below the Chinese standard. On the other hand, we simulated the consequences of hydrogen leak from the 45 MPa hydrogen storage vessels by the physic module of HyRAM and obtained the ranges of plume dispersion, jet fire, radiative heat flux, and unconfined overpressure. We suggest targeted preventive measures and safety distance to provide references for hydrogen fueling stations’ safe construction and operation.

To accurate modulate the phase of the incoming light, the backplane aberration of a spatial light modulator (SLM) needs to be measured and compensated. In this paper, we develop an interferometric method to calibrate the static aberration. In our method, a Michelson interferometer was constructed and the SLM itself was used to produce the random phase shift that we need. In addition, the phase demodulation method based on matrix VU factorization (VU) and phase unwrapping algorithm based on derivative Zernike polynomial fitting (DZPT) are adopted to get the phase profile of the static aberration. Experimental result shows that our proposed method can get a pretty good compensation result.

Despite the great progress made by deep neural networks in the semantic segmentation task, traditional neural-network-based methods typically suffer from a shortage of large amounts of pixel-level annotations. Recent progress in few-shot semantic segmentation tackles the issue by only a few pixel-level annotated examples. However, these few-shot approaches cannot easily be applied to multi-way or weak an-notation settings. In this paper, we advance the few-shot segmentation paradigm towards a scenario where image-level an-notations are available to help the training process of a few pixel-level annotations. Our key idea is to learn a better prototype representation of the class by fusing the knowledge from the image-level labeled data. Specifically, we propose a new framework, called PAIA, to learn the class prototype representation in a metric space by integrating image-level annotations. Furthermore, by considering the uncertainty of pseudo-masks, a distilled soft masked average pooling strategy is designed to handle distractions in image-level annotations. Extensive empirical results on two datasets show superior performance of PAIA.

An improved and practical method is reported here for accessing 3′,4′,5′-trifluoro-[1,1′-biphenyl]-2-amine (1), a key intermediate for Fluxapyroxad. The overall yield for the preparation of 1 was 7...

The presence of data noise and corruptions has recently invoked increasing attention on robust least-squares regression ( RLSR ), which addresses this fundamental problem that learns reliable regression coefficients when response variables can be arbitrarily corrupted. Until now, the following important challenges could not be handled concurrently: (1) rigorous recovery guarantee of regression coefficients, (2) difficulty in estimating the corruption ratio parameter, and (3) scaling to massive datasets. This article proposes a novel Robust regression algorithm via Heuristic Corruption Thresholding ( RHCT ) that concurrently addresses all the above challenges. Specifically, the algorithm alternately optimizes the regression coefficients and estimates the optimal uncorrupted set via heuristic thresholding without a pre-defined corruption ratio parameter until its convergence. Moreover, to improve the efficiency of corruption estimation in large-scale data, a Robust regression algorithm via Adaptive Corruption Thresholding ( RACT ) is proposed to determine the size of the uncorrupted set in a novel adaptive search method without iterating data samples exhaustively. In addition, we prove that our algorithms benefit from strong guarantees analogous to those of state-of-the-art methods in terms of convergence rates and recovery guarantees. Extensive experiments demonstrate that the effectiveness of our new methods is superior to that of existing methods in the recovery of both regression coefficients and uncorrupted sets, with very competitive efficiency.

9136 Background: Lung cancer organoids (LCOs) were expected to be the potential precision medicine approach for clinical response prediction. However, the clinical applications of both tissue and malignant serous effusions (MSE) derived LCOs were rarely reported. Our previous work demonstrated that MSE-derived LCOs maintain the genomic signature of the original tumor. In this study, we aimed to create LCOs using tissue or MSE, then validate the reliability of the model by comparing LCOs and their origin from the pathological and molecular levels. Furthermore, drug sensitivity tests of LCOs were also performed to evaluate the feasibility of LCO drug test as an approach for personalized medicine. Methods: Primary or metastatic tumor tissues were obtained from advanced lung cancer patients through core biopsy or surgically resected biopsy at the Guangdong Provincial People’s Hospital. MSEs were also collected. LCOs were generated from the obtained tissue and MSE, and the pathological features and genomic profiles were verified by analyzing the consistency with their origin. Then, the drug sensitivity scheme was formulated to follow the principles of clinical medication. In addition, proteomics analysis by 4D LC-MS/MS was also performed to analysis the molecular details of combinational therapy. Results: In our study, we generated 213 LCOs from 106 patients, mainly from MSE. The success rate to generate LCOs derived from MSE was 81.4% (131/161). The concordance rate of pathological phenotypes of LCOs samples verified by immunohistochemistry with clinical samples was 75% (63/84). In our cohort, LCO based drug sensitivity tests (LCO-DST) of targeted therapies were performed to predict the tumor response, and the AUC value of ROC analysis of osimertinib in EGFR-mutant adenocarcinoma reached 0.94(LCOs samples = 15, p= 0.0047). There were 2 patients with advanced lung adenocarcinoma, one with de novo EGFR mutation /MET amplification and the other with EGFR mutation combined with acquired RET fusion. The results of LCO based drug tests of 2 patients showed that combined targeted therapy (osimertinib plus savolitinib/cabozantinib) showed high tumor inhibition rate validated in clinical treatment and made differences. Then, 4D label-free high through-put proteomic analysis was performed in the patient with EGFR mutation and acquired RET fusion, demonstrating caspase 3 increased dramatically in combination of osimertinib and BLU-667 and the downstream proteins of EGFR and RET were down-regulated. Conclusions: LCOs derived from MSE faithfully reflected the pathological and genomic features of their original patients. The LCOs based drug test results are remarkably consistent with the tumor response. These results suggested the important prospects of LCO as an in vitro model for lung cancer precision medicine.

e21021 Background: BRAF variants were reported resistant mechanisms to epidermal growth factor receptor (EGFR)-tyrosine kinase inhibitors (TKIs) in EGFR-mutant non-small cell lung cancer (NSCLC). However, concomitant somatic variations other than BRAF variants and efficacy of subsequent treatment remained unclear. Methods: From October 2016 to May 2020, advanced NSCLC patients who underwent next-generation sequencing and were detected with co-mutation of BRAF and EGFR activating mutations are retrospectively included. Since June 2020, EGFR-mutant patients with acquired BRAF V600E after progression from previous Osimertinib are prospectively arranged to explore efficacy of the EGFR-BRAF co-inhibition. Results: Fifty-eight patients were retrospectively identified and five patients were prospectively included. BRAF variants was acquired after a median time of 22.7 months from initial diagnosis. Variations of TP53, PIK3CA, RB1, MET, LRP1B, APC, CDKN2A, MYC, ERBB2 and SMAD4 were over 10%, which were enriched in cell cycle/p53 pathway, EGFR downstream and bypass pathway. Subsequently, median progression-free survival was 5.0 months for chemotherapy and 2.1 months for TKI treatment without targeting both EGFR and BRAF respectively (p = 0.019). Osimertinib plus vemurafenib (n = 4) or dabrafenib + trametinib (n = 1) was prospectively administrated in five patients. Median PFS was 7.8 months. Grade 3 rash was observed in one patient. Upon disease progression, activation of RAS signaling was observed. Conclusions: Variations of EGFR downstream or bypass pathway were also frequent in patient with co-mutation of EGFR and BRAF. Efficacy of subsequent TKI without targeting both EGFR and BRAF was inferior to chemotherapy. EGFR-BRAF co-inhibition showed improved efficacy. More treatment strategy should be explored in the future.

9032 Background: Immunotherapy has become standard therapy for untreated advanced NSCLC. However, no direct comparison between anti-PD-1 inhibitors has been reported. Methods: CTONG1901 is an open label, randomized, phase II clinical trial to compare sintilimab and pembrolizumab in monotherapy or combination with chemotherapy for advanced NSCLC at first-line setting. The primary endpoint was objective response rate (ORR). Patient without EGFR and ALK alteration were enrolled. Patients with PD-L1 tumor proportion score (TPS) ≥50% were randomly to receive sintilimab (A) or pembrolizumab (B) ; and with TPSst stage. When ≥4 patients achieve partial response (PR) in sintilimab arms, the study will enter into 2nd stage and the sample size will be calculated based on the ORR results of the 1st stage. Results: The ORR was 57.1% in sintilimab and 33.3% in pembrolizumab arms at the 1st stage. The study successfully entered into the 2nd stage. 48 additional patients should be enrolled after calculation. When 15 PR in sintilimab arms achieved, the primary endpoint will be reached. From Mar. 2020 to Jan. 2022, 71 patients were screened and 68 patients were enrolled in two stages. Histologic subtypes and brain metastasis were well balanced between arms. As of Dec. 31st 2021, the median follow-up was 5.6 months. The confirmed ORR was 45.5% (15/33) in sintilimab vs. 28.6% (10/35) in pembrolizumab arms (A vs. B: 30.8% [4/13] vs. 28.6% [4/14]; C vs. D: 55.0% [11/20] vs. 28.6% [6/21]). Unconfirmed ORR was 57.6% vs. 42.9% and disease control rate (DCR) was 87.9% vs. 91.4% in sintilimab and pembrolizumab arms. The primary endpoint was reached. Survival data was immature. Any grade and 3-4 grade treatment-related adverse events (TRAEs) were comparable in sintilimab and pembrolizumab arms (Table). Conclusions: This is the first head-to-head phase II study to directly compare two anti-PD-1 antibodies as first-line treatment in advanced NSCLC. The result suggested comparable tumor response and similar safety profile between sintilimab and pembrolizumab. Clinical trial information: NCT04252365. [Table: see text]

8578 Background: IMpower133 showed benefits in both progression-free survival (PFS) and overall survival (OS) of etoposide/carboplatin plus atezolizumab (ECT) regimen in extensive-stage small-cell lung cancer (ES-SCLC), but the absolute benefit was limited. Previous studies have classified SCLC patients by RNA-seq clustering analysis to explore the dominant population for treatment, but was difficult for clinical application. We aimed to explore whether expressive status of Yes-Associated Protein 1 (YAP-1) can screen dominant population of immunotherapy among ES-SCLC patients. Methods: We selected ES-SCLC patients treated in our hospital from Jan, 2018 to Jul, 2021, and enrolled 21 patients with ES-SCLC received ECT regimen whose formaldehyde-fixed, paraffin-embedded sample was reachable. Assessments of complete remission (CR), partial remission (PR), disease stable (SD) and progressive disease (PD) were according to the efficacy evaluation criteria of solid tumor (RECIST) version 1.1. Immunohistochemistry (IHC) of YAP-1 (ET1608-30, 1/100) was conducted. The H-score was calculated by IHC Profiler. All statistical analyses were evaluated using SPSS 22.0, X-tile 3.6.1, and Excel. P values < 0.05 were considered statistically significant. Results: Baseline information was provided in table. The median H-score of responders (CR/PR patients) and non-responders were 13.97 (95%CI: 8.97-16.30) and 23.72 (95%CI: 8.13-75.40) that were significantly different (P＜0.05). H-score and PFS showed negative correlation by spearman (r = -0.603). Patients were divided into two groups as low expression group (H-score ≤25.00, n = 16) and high expression group (H-score ＞25.00, n = 5) according to the cut-off value of H-score. The median PFS of these two groups were 7.1m (95%CI: 2.6-11.6m) and 3.4m (95%CI: 0.9-5.9m), respectively. K-M curves of PFS were significantly different (P＜0.05). Conclusions: Our preliminary results have indicated a potential efficacy predictive value of YAP-1 protein. And the expression level of YAP-1 protein was negatively correlated with efficacy of ECT in ES-SCLC patients. [Table: see text]

Measuring document similarity plays an important role in natural language processing tasks. Most existing document similarity approaches suffer from the information gap caused by context and vocabulary mismatches when comparing varying-length texts. In this paper, we propose an unsupervised concept representation learning approach to address the above issues. Specifically, we propose a novel Concept Generation Network (CGNet) to learn concept representations from the perspective of the entire text corpus. Moreover, a concept-based document matching method is proposed to leverage advances in the recognition of local phrase features and corpus-level concept features. Extensive experiments on real-world data sets demonstrate that new method can achieve a considerable improvement in comparing length-varying texts. In particular, our model achieved 6.5% better F1 Score compared to the best of the baseline models for a concept-project benchmark dataset.

Multi-target regression has recently drawn increasing attention in the machine learning community due to its capability of simultaneously predicting multiple continuous target variables based on a given set of input features. Jointly handling the inter-target correlations and input-output relationships is very challenging. That task becomes even more intricate in the presence of correlated data corruption. We observe that traditional robust methods can hardly deal with several emerging challenges, including 1) presence of correlated corruption among targets in the datasets, 2) difficulty in estimating the data corruption ratio, and 3) scalability to massive datasets. This paper proposes a novel approach that addresses all the above challenges by developing a distributed robust regression algorithm. Specifically, the algorithm optimizes regression coefficients of each target in parallel with a heuristically estimated corruption ratio and then consolidates the uncorrupted set in two strategies: global consensus and majority voting. Also, we prove that our algorithm benefits from strong guarantees in terms of convergence rates and coefficient recovery, which can be applied as a generic framework for robust regression problem with correlated corruption property. Extensive experiments on synthetic and real-world datasets demonstrate that our algorithm is superior to existing methods in both effectiveness and efficiency.

Small intestinal neuroendocrine carcinomas (SI NECs) are diagnosed very rarely, and the prognosis is extremely poor due to the metastatic disease of most patients at the time of diagnosis. This study aimed to establish nomogram models for prognostic evaluation of SI NEC in both overall survival (OS) and cancer-specific survival (CSS). Patients diagnosed with SI NEC between 2010 and 2015 were retrieved from the Surveillance, Epidemiology, and End Results (SEER) database and further randomly divided into the training and validating cohorts at a ratio of 7:3. Univariate and multivariate cox analysis was conducted to determine significant variables for construction of nomogram. The performance of the nomogram models were then assessed by concordance index (C-index), calibration plot and the area receiver operating characteristic (ROC) curve (AUC). A total of 1110 patients were retrospectively selected from the SEER database. Multivariate models revealed that age, tumor grade, American Joint Committee for Cancer (AJCC) stage, surgery and chemotherapy all showed a significant association with OS and CSS. The discrimination of nomogram for OS prediction was superior to that of the 7th AJCC Tumor-Node-Metastasis (TNM) staging system (C-index = 0.798, 95% CI, 0.762 - 0.833 vs 0.623, 95% CI, 0.580 - 0.666, P < 0.001). Similar results were also observed in CSS nomogram. Well-corresponded calibration plots were noticed using the nomograms. The comparisons of AUC values showed that the established nomograms exhibited better discrimination power than 7th TNM staging system for OS and CSS prediction. In conclusion, we have successfully established novel nomograms for predicting OS and CSS in patients with SI NEC, which can assist clinicians in making predictions about individual patient survival and provide improved treatment strategies.

Background: This study aimed to develop and validate nomograms for predicting overall survival (OS) and cancer-specific survival (CSS) in small intestinal gastrointestinal stromal tumors (SI GISTs). Methods: Patients diagnosed with SI GISTs were retrieved from the Surveillance, Epidemiology, and End Results (SEER) database and further randomly divided into training and validating cohorts. Univariate and multivariate Cox analyses were conducted in the training set to determine independent prognostic factors to build nomograms for predicting 3- and 5-year OS and CSS. The performance of the nomograms was assessed by using the concordance index (C-index), calibration plot, and the area under the receiver operating characteristic curve (AUC). Results: Data of a total of 776 patients with SI GISTs were retrospectively collected from the SEER database. The OS nomogram was constructed based on age, surgery, imatinib treatment, and American Joint Committee for Cancer (AJCC) stage, while the CSS nomogram incorporated age, surgery, tumor grade, and AJCC stage. In the training set, the C-index for the OS nomogram was 0.773 (95% confidence interval [95% CI]: 0.722–0.824) and for the CSS nomogram 0.806 (95% CI: 0.757–0.855). In the internal validation cohort, the C-index for the OS nomogram was 0.741, while for the CSS nomogram, it was 0.819. Well-corresponded calibration plots both in OS and CSS nomogram models were noticed. The comparisons of AUC values showed that the established nomograms exhibited superior discrimination power than the 7th Tumor-Node-Metastasis staging system. Conclusion: Our nomogram can effectively predict 3- and 5-year OS and CSS in patients with SI GISTs, and its use can help improve the accuracy of personalized survival prediction and facilitate to provide constructive therapeutic suggestions.

Myeloid-derived suppressor cells (MDSCs) suppress antitumor immune activities and facilitate cancer progression. Although the concept of immunosuppressive MDSCs is well established, the mechanism that MDSCs regulate non-small cell lung cancer (NSCLC) progression through the paracrine signals is still lacking. Here, we reported that the infiltration of MDSCs within NSCLC tissues was associated with the progression of cancer status, and was positively correlated with the Patient-derived xenograft model establishment, and poor patient prognosis. Intratumoral MDSCs directly promoted NSCLC metastasis and highly expressed chemokines that promote NSCLC cells invasion, including CCL11. CCL11 was capable of activating the AKT and ERK signaling pathways to promote NSCLC metastasis through the epithelial-mesenchymal transition (EMT) process. Moreover, high expression of CCL11 was associated with a poor prognosis in lung cancer as well as other types of cancer. Our findings underscore that MDSCs produce CCL11 to promote NSCLC metastasis via activation of ERK and AKT signaling and induction of EMT, suggesting that the MDSCs-CCL11-ERK/AKT-EMT axis contains potential targets for NSCLC metastasis treatment.