Oral_voting_transfer: classification of oral microorganisms' function proteins with voting transfer model

Introduction

The oral microbial group typically represents the human body's highly complex microbial group ecosystem. Oral microorganisms take part in human diseases, including Oral cavity inflammation, mucosal disease, periodontal disease, tooth decay, and oral cancer. On the other hand, oral microbes can also cause endocrine disorders, digestive function, and nerve function disorders, such as diabetes, digestive system diseases, and Alzheimer's disease. It was noted that the proteins of oral microbes play significant roles in these serious diseases. Having a good knowledge of oral microbes can be helpful in analyzing the procession of related diseases. Moreover, the high-dimensional features and imbalanced data lead to the complexity of oral microbial issues, which can hardly be solved with traditional experimental methods.

Methods

To deal with these challenges, we proposed a novel method, which is oral_voting_transfer, to deal with such classification issues in the field of oral microorganisms. Such a method employed three features to classify the five oral microorganisms, including Streptococcus mutans, Staphylococcus aureus, abiotrophy adjacent, bifidobacterial, and Capnocytophaga. Firstly, we utilized the highly effective model, which successfully classifies the organelle's proteins and transfers to deal with the oral microorganisms. And then, some classification methods can be treated as the local classifiers in this work. Finally, the results are voting from the transfer classifiers and the voting ones.

Results and discussion

The proposed method achieved the well performances in the five oral microorganisms. The oral_voting_transfer is a standalone tool, and all its source codes are publicly available at https://github.com/baowz12345/voting_transfer.

Epicardial and pericoronary adipose tissue and coronary plaque burden in patients with Cushing's syndrome: a propensity score-matched study

PURPOSE

To assess coronary inflammation by measuring the volume and density of the epicardial adipose tissue (EAT), perivascular fat attenuation index (FAI) and coronary plaque burden in patients with Cushing's syndrome (CS) based on coronary computed tomography angiography (CCTA).

METHODS

This study included 29 patients with CS and 58 matched patients without CS who underwent CCTA. The EAT volume, EAT density, FAI and coronary plaque burden were measured. The high-risk plaque (HRP) was also evaluated. CS duration from diagnosis, 24-h urinary free cortisol (UFC), and abdominal visceral adipose tissue volume (VAT) of CS patients were recorded.

RESULTS

The CS group had higher EAT volume (146.9 [115.4, 184.2] vs. 119.6 [69.0, 147.1] mL, P = 0.006), lower EAT density (- 78.79 ± 5.89 vs. - 75.98 ± 6.03 HU, P = 0.042), lower FAI (- 84.0 ± 8.92 vs. - 79.40 ± 10.04 HU, P = 0.038), higher total plaque volume (88.81 [36.26, 522.5] vs. 44.45 [0, 198.16] mL, P = 0.010) and more HRP plaques (7.3% vs. 1.8%, P = 0.026) than the controls. The multivariate analysis suggested that CS itself (β [95% CI], 29.233 [10.436, 48.03], P = 0.014), CS duration (β [95% CI], 0.176 [0.185, 4.242], P = 0.033), and UFC (β [95% CI], 0.197 [1.803, 19.719], P = 0.019) were strongly associated with EAT volume but not EAT density, and EAT volume (β [95% CI] - 0.037[- 0.058, - 0.016], P = 0.001) not CS was strongly associated with EAT density. EAT volume, FAI and plaque burden increased (all P < 0.05) in 6 CS patients with follow-up CCTA. The EAT volume had a moderate correlation with abdominal VAT volume (r = 0.526, P = 0.008) in CS patients.

CONCLUSIONS

Patients with CS have higher EAT volume and coronary plaque burden but less inflammation as detected by EAT density and FAI. The EAT density is associated with EAT volume but not CS itself.

Identify Diabetes-related Targets based on ForgeNet_GPC

BACKGROUND

Research on potential therapeutic targets and new mechanisms of action can greatly improve the efficiency of new drug development.

AIMS

Polygenic genetic diseases, such as diabetes, are caused by the interaction of multiple gene loci and environmental factors.

OBJECTIVE

In this study, a disease target identification algorithm based on protein recognition is proposed.

METHODS

In this method, the related and unrelated targets are collected from literature databases for treating diabetes. The transcribed proteins corresponding to each target are queried in order to construct a protein dataset. Six protein feature extraction algorithms (AAC, CKSAAGP, DDE, DPC, GAAP, and TPC) are utilized to obtain the feature vectors of each protein, which are merged into the full feature vectors.

RESULTS

A novel classifier (forgeNet_GPC) based on forgeNet and Gaussian process classifier (GPC) is proposed to classify the proteins.

CONCLUSION

In forgeNet_GPC, forgeNet is utilized to select the important features, and GPC is utilized to solve the classification problem. The experimental results reveal that forgeNet_GPC performs better than 22 classifiers in terms of ROC-AUC, PR-AUC, MCC, Youden Index, and Kappa.

Application of Network Pharmacology and Molecular Docking to Explore the Mechanism of Danggui Liuhuang Tang against Hyperthyroidism

INTRODUCTION

To investigate the mechanism of Danggui Liuhuang Tang (DGLHT) in the treatment of hyperthyroidism (HT), we explored the multi-component, multi-target, and multi-pathway mechanism based on the network pharmacology method of traditional Chinese medicine.

OBJECTIVE

Using network pharmacology and molecular docking, the effective components, core targets, and critical pathways of DGLHT in the therapy of HT were investigated. The mechanism of DGLHT in the treatment of HT is discussed in this work, which also offers a scientific foundation for further research into the process.

METHODS

To take DGLHT into the blood components as the research object, we used GeneCards, Drungbank, Therapeutic Target Database (TTD), Online Mendelian Inheritance in Man (OMIM), Pharmacogenetics and Pharmacogenomics Knowledge Base (PharmGKB), and other databases to predict the potential target of the components. Then, it was integrated with the predicted targets of HT disease to obtain the potential targets of DGLHT in the treatment of HT. We used String database and Cytoscape software for protein-protein interaction network (PPI) construction, and DAVID platform for Gene Ontology (GO) analysis and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway annotation, the Cytoscape software was used to construct a "component-target-pathway" network; the AutoDock Vina platform was used to conduct molecular docking between the blood entry components and key targets.

RESULTS

According to the analysis, a total of 93 active ingredients, 348 disease-related targets, and 36 potential targets were screened out. Among them, key targets such as MAPK1, CCND1, AKT1, and TNF exert curative effects, and the main pathways are the HIF-1 signaling pathway, FoxO signaling pathway, Chemokine signaling pathway, TNF signaling pathway, Toll-like receptor signaling pathway, T cell receptor signaling pathway, Jak-STAT signaling pathway, and other pathways. Molecular docking results verified the interaction between active ingredients and key targets, among which rustication and quercetin had high docking affinity with key target proteins MAPK1 and CCND1.

CONCLUSION

This study preliminary revealed that DGLHT has the characteristics of multi-component, multi-target, and multi-pathway in the treatment of HT, and it established a scientific foundation for a more detailed investigation of DGLHT's molecular mechanism in the treatment of HT.

Corrigendum: Alzheimer-compound identification based on data fusion and forgeNet_SVM

[This corrects the article DOI: 10.3389/fnagi.2022.931729.].

Golgi_DF: Golgi proteins classification with deep forest

Introduction

Golgi is one of the components of the inner membrane system in eukaryotic cells. Its main function is to send the proteins involved in the synthesis of endoplasmic reticulum to specific parts of cells or secrete them outside cells. It can be seen that Golgi is an important organelle for eukaryotic cells to synthesize proteins. Golgi disorders can cause various neurodegenerative and genetic diseases, and the accurate classification of Golgi proteins is helpful to develop corresponding therapeutic drugs.

Methods

This paper proposed a novel Golgi proteins classification method, which is Golgi_DF with the deep forest algorithm. Firstly, the classified proteins method can be converted the vector features containing various information. Secondly, the synthetic minority oversampling technique (SMOTE) is utilized to deal with the classified samples. Next, the Light GBM method is utilized to feature reduction. Meanwhile, the features can be utilized in the penultimate dense layer. Therefore, the reconstructed features can be classified with the deep forest algorithm.

Results

In Golgi_DF, this method can be utilized to select the important features and identify Golgi proteins. Experiments show that the well-performance than the other art-of-the state methods. Golgi_DF as a standalone tools, all its source codes publicly available at https://github.com/baowz12345/golgiDF.

Discussion

Golgi_DF employed reconstructed feature to classify the Golgi proteins. Such method may achieve more available features among the UniRep features.

PGRNIG: novel parallel gene regulatory network identification algorithm based on GPU

Molecular biology has revealed that complex life phenomena can be treated as the result of many gene interactions. Investigating these interactions and understanding the intrinsic mechanisms of biological systems using gene expression data have attracted a lot of attention. As a typical gene regulatory network (GRN) inference method, the S-system has been utilized to deal with small-scale network identification. However, it is extremely difficult to optimize it to infer medium-to-large networks. This paper proposes a novel parallel swarm intelligent algorithm, PGRNIG, to optimize the parameters of the S-system. We employed the clone selection strategy to improve the whale optimization algorithm (CWOA). To enhance the time efficiency of CWOA optimization, we utilized a parallel CWOA (PCWOA) based on the compute unified device architecture (CUDA) platform. Decomposition strategy and L1 regularization were utilized to reduce the search space and complexity of GRN inference. We applied the PGRNIG algorithm on three synthetic datasets and two real time-series expression datasets of the species of Escherichia coli and Saccharomyces cerevisiae. Experimental results show that PGRNIG could infer the gene regulatory network more accurately than other state-of-the-art methods with a convincing computational speed-up. Our findings show that CWOA and PCWOA have faster convergence performances than WOA.

Alzheimer-Compound Identification Based on Data Fusion and forgeNet_SVM

Rapid screening and identification of potential candidate compounds are very important to understand the mechanism of drugs for the treatment of Alzheimer's disease (AD) and greatly promote the development of new drugs. In order to greatly improve the success rate of screening and reduce the cost and workload of research and development, this study proposes a novel Alzheimer-related compound identification algorithm namely forgeNet_SVM. First, Alzheimer related and unrelated compounds are collected using the data mining method from the literature databases. Three molecular descriptors (ECFP6, MACCS, and RDKit) are utilized to obtain the feature sets of compounds, which are fused into the all_feature set. The all_feature set is input to forgeNet_SVM, in which forgeNet is utilized to provide the importance of each feature and select the important features for feature extraction. The selected features are input to support vector machines (SVM) algorithm to identify the new compounds in Traditional Chinese Medicine (TCM) prescription. The experiment results show that the selected feature set performs better than the all_feature set and three single feature sets (ECFP6, MACCS, and RDKit). The performances of TPR, FPR, Precision, Specificity, F1, and AUC reveal that forgeNet_SVM could identify more accurately Alzheimer-related compounds than other classical classifiers.

Single_cell_GRN: gene regulatory network identification based on supervised learning method and Single-cell RNA-seq data

Single-cell RNA-seq overcomes the shortcomings of conventional transcriptome sequencing technology and could provide a powerful tool for distinguishing the transcriptome characteristics of various cell types in biological tissues, and comprehensively revealing the heterogeneity of gene expression between cells. Many Intelligent Computing methods have been presented to infer gene regulatory network (GRN) with single-cell RNA-seq data. In this paper, we investigate the performances of seven classifiers including support vector machine (SVM), random forest (RF), Naive Bayesian (NB), GBDT, logical regression (LR), decision tree (DT) and K-Nearest Neighbor (KNN) for solving the binary classification problems of GRN inference with single-cell RNA-seq data (Single_cell_GRN). In SVM, three different kernel functions (linear, polynomial and radial basis function) are utilized, respectively. Three real single-cell RNA-seq datasets from mouse and human are utilized. The experiment results prove that in most cases supervised learning methods (SVM, RF, NB, GBDT, LR, DT and KNN) perform better than unsupervised learning method (GENIE3) in terms of AUC. SVM, RF and KNN have the better performances than other four classifiers. In SVM, linear and polynomial kernels are more fit to model single-cell RNA-seq data.

Disease-Ligand Identification Based on Flexible Neural Tree

In order to screen the disease-related compounds of a traditional Chinese medicine prescription in network pharmacology research accurately, a new virtual screening method based on flexible neural tree (FNT) model, hybrid evolutionary method and negative sample selection algorithm is proposed. A novel hybrid evolutionary algorithm based on the Grammar-guided genetic programming and salp swarm algorithm is proposed to infer the optimal FNT. According to hypertension, diabetes, and Corona Virus Disease 2019, disease-related compounds are collected from the up-to-date literatures. The unrelated compounds are chosen by negative sample selection algorithm. ECFP6, MACCS, Macrocycle, and RDKit are utilized to numerically characterize the chemical structure of each compound collected, respectively. The experiment results show that our proposed method performs better than classical classifiers [Support Vector Machine (SVM), random forest (RF), AdaBoost, decision tree (DT), Gradient Boosting Decision Tree (GBDT), KNN, logic regression (LR), and Naive Bayes (NB)], up-to-date classifier (gcForest), and deep learning method (forgeNet) in terms of AUC, ROC, TPR, FPR, Precision, Specificity, and F1. MACCS method is suitable for the maximum number of classifiers. All methods perform poorly with ECFP6 molecular descriptor.

A survey on protein–DNA-binding sites in computational biology

Transcription factors are important cellular components of the process of gene expression control. Transcription factor binding sites are locations where transcription factors specifically recognize DNA sequences, targeting gene-specific regions and recruiting transcription factors or chromatin regulators to fine-tune spatiotemporal gene regulation. As the common proteins, transcription factors play a meaningful role in life-related activities. In the face of the increase in the protein sequence, it is urgent how to predict the structure and function of the protein effectively. At present, protein–DNA-binding site prediction methods are based on traditional machine learning algorithms and deep learning algorithms. In the early stage, we usually used the development method based on traditional machine learning algorithm to predict protein–DNA-binding sites. In recent years, methods based on deep learning to predict protein–DNA-binding sites from sequence data have achieved remarkable success. Various statistical and machine learning methods used to predict the function of DNA-binding proteins have been proposed and continuously improved. Existing deep learning methods for predicting protein–DNA-binding sites can be roughly divided into three categories: convolutional neural network (CNN), recursive neural network (RNN) and hybrid neural network based on CNN–RNN. The purpose of this review is to provide an overview of the computational and experimental methods applied in the field of protein–DNA-binding site prediction today. This paper introduces the methods of traditional machine learning and deep learning in protein–DNA-binding site prediction from the aspects of data processing characteristics of existing learning frameworks and differences between basic learning model frameworks. Our existing methods are relatively simple compared with natural language processing, computational vision, computer graphics and other fields. Therefore, the summary of existing protein–DNA-binding site prediction methods will help researchers better understand this field.

OP0260 RESPONSIVENESS OF A COMBINED POWER DOPPLER AND GREYSCALE ULTRASOUND SCORE FOR ASSESSING SYNOVITIS AT JOINT LEVEL IN PSORIATIC ARTHRITIS PATIENTS WITH INADEQUATE RESPONSE TO csDMARDs: DATA FROM THE ULTIMATE TRIAL

Background

Power Doppler ultrasound (PDUS) is a sensitive non-invasive imaging tool that allows the visualisation of articular and periarticular inflammation in patients with psoriatic arthritis (PsA).1 ULTIMATE (NCT02662985) was the first large randomised clinical trial that showed the responsiveness of the Global OMERACT-EULAR ultrasound synovitis score (GLOESS) in PsA and confirmed the rapid and continued benefits of secukinumab through 52 weeks.2,3

Objectives

To report the distribution of ultrasound-detected synovitis at joint level, by degree of severity at baseline and over time, and the contribution of each core component of GLOESS, synovial hypertrophy (SH) by greyscale (GS; B-mode) and power Doppler (PD) signal, to responsiveness.3

Methods

This was a 52-week study with a 12-week double-blind, placebo-controlled treatment period followed by a 12-week open-label period and a 6-month open-label extension secukinumab treatment period.3 The number of joints with synovitis measured by GLOESS2 was assessed up to Week 52. The assessments included distribution of synovitis based on composite PDUS score across 24 pairs of joints (with worse score of the pair of the joints used) by grade of severity (0-3) and change from baseline to Week 52 in each core component of GLOESS.3-5 Data are presented as observed.

Results

A total of 166 patients (mean age, 46.7 years; males, 45.2%) were enrolled, of which 90% (75/83) of secukinumab and 83% (69/83) of placebo-secukinumab participants completed 52 weeks. The mean (SD) number of PDUS detected synovitis at baseline was 9.2 (4.9) and 10.2 (5.2) in the secukinumab and placebo group, respectively. The most frequent locations with synovitis at baseline were: wrist, metatarsophalangeal (MTP), knees and metacarpophalangeal (MCP) joints (Table 1). An early and continued improvement in GLOESS was observed in both secukinumab and placebo-secukinumab groups after switching to active therapy, as previously reported at Week 12 and through Week 52.2,3 Among the two core components of GLOESS, SH was mainly responsible for the change in GLOESS from baseline to Week 52, in contrast with PD signal in this dataset. The distribution of synovitis by grade of severity showed that MTP joints, wrist, knee, MCP1/2 and tibiotalar joints mostly contributed to the composite PDUS at Week 12 (Figure 1). Similar patterns were observed through 52 weeks.

Table 1.

Proportion of patients with PDUS detected synovitis at baseline*

Synovitis joint, data presented as n (%)Secukinumab (N=83)Placebo (N=83)Wrist66 (80)66 (80)MTP256 (68)65 (78)MTP158 (70)60 (72)MTP352 (63)60 (72)MTP446 (55)59 (71)Knee50 (60)47 (57.)MCP136 (43)52 (63)MCP235 (42)41 (49)MTP530 (36)41 (49)

*Data for top nine pairs of joints with most frequently detected power Doppler ultrasound (PDUS) synovitis are presented here. Synovitis was scored by a OMERACT-EULAR synovitis composite score >0 for each paired joint (irrespective of right or left side). The OMERACT-EULAR composite PDUS score (for individual joints) ranged from 0 to 3 and was composed of the two core components synovial hypertrophy and power Doppler.

N, total number of randomised patients; n, number of evaluable patients

Conclusion

The distribution of synovitis at baseline reflected a predominance of small joints (feet and hands) and large joints (wrist and knee) and were mostly responsive to secukinumab over time in the ULTIMATE trial. Synovial hypertrophy was the most responsive core component of GLOESS driving an early and continued reduction of synovitis with secukinumab through Week 52. This finding could be useful to select a restricted number of joints in future ultrasound trials in PsA.

References

[1]D’Agostino MA and Coates LC. J Rheumatol. 2019;46:337–9.

[2]D’Agostino MA et al. Arthritis Rheumatol. 2021;73(10).

[3]D’Agostino MA, et al. Rheumatology (Oxford) 2021;keab628.

[4]D’Agostino MA and Coates LC. RMD Open 2017;3:e000428.

[5]Uson J, et al. Rheumatol Clin. 2018;14:27–35.

Disclosure of Interests

Maria-Antonietta D’Agostino Speakers bureau: Sanofi, Novartis, BMS, Janssen, Celgene, Roche, AbbVie, UCB, and Eli Lilly, Consultant of: Sanofi, Novartis, BMS, Janssen, Celgene, Roche, AbbVie, UCB, and Eli Lilly, Maarten Boers Consultant of: Novartis, Corine Gaillez Shareholder of: Novartis and BMS, Employee of: Novartis, Carlos Gamez: None declared, LUCIO VENTURA: None declared, Javier Rosa Speakers bureau: Abbvie, Pfizer, Lilly, Janssen, Novartis and BMS, Ilaria Padovano: None declared, Peter Mandl Speakers bureau: AbbVie, BMS, Celgene, Janssen, Lilly, MSD, Novartis, Roche and UCB, Grant/research support from: AbbVie, BMS, Celgene, Janssen, Lilly, MSD, Novartis, Roche and UCB, Arnd Kleyer Speakers bureau: Abbvie, Lilly, Novartis, MEDAC; Janssen, Consultant of: Abbvie, Lilly, UCB, Novartis, BMS, Sanofi, Galapagos, Catherine Bakewell Speakers bureau: AbbVie, Novartis, Pfizer, Janssen, UCB, and Sanofi Genzyme/Regeneron, Consultant of: AbbVie, Novartis, Pfizer, Janssen, UCB, and Sanofi Genzyme/Regeneron, Weibin Bao Shareholder of: Novartis, Employee of: Novartis, Punit Goyanka Employee of: Novartis, Philip G Conaghan Speakers bureau: AbbVie, Amgen, AstraZeneca, Eli Lilly, Galapagos, Gilead, Novartis, Pfizer and UCB, Consultant of: AbbVie, Amgen, AstraZeneca, Eli Lilly, Galapagos, Gilead, Novartis, Pfizer and UCB, Georg Schett Speakers bureau: AbbVie, BMS, Celgene, Janssen, Lilly, Novartis, Roche and UCB

OP0294 REDUCED JOINT SYNOVITIS ASSESSMENT VERSUS THE GLOBAL EULAR OMERACT SYNOVITIS SCORE (GLOESS) TO PREDICT THE RESPONSE TO SECUKINUMAB IN PATIENTS WITH ACTIVE PSORIATIC ARTHRITIS AND INADEQUATE RESPONSE TO CONVENTIONAL DISEASE-MODIFYING ANTI-RHEUMATIC DRUGS: EXPLORATORY RESULTS FROM THE ULTIMATE TRIAL

Background

The combined use of B-mode ultrasound (US) and Power Doppler (PD; combination termed as PDUS) allows visualisation of morphological and pathophysiological changes of the synovium. ULTIMATE (NCT02662985) was the first large, randomised, double-blind, placebo-controlled PDUS phase IIIb study in psoriatic arthritis (PsA), to demonstrate that Global OMERACT EULAR Synovitis Score (GLOESS), a US score at patient level, was sensitive to detect the early and continuous decrease in synovitis in a multicenter setting using different US devices and examiners.1 However, the US assessment for GLOESS was time-consuming owing to the number of joints assessed.

Objectives

To investigate the value of various reduced joint sets to predict the validated GLOESS score.

Methods

ULTIMATE was a 52-week study with a 12-week double-blind, placebo-controlled period followed by 12-week open-label (OL) treatment and 6-month OL extension period.1 In the ULTIMATE trial, GLOESS for the 24 paired joints was calculated, with a potential score ranging between 0 to 144.1 A Spearman’s rank correlation matrix and a Cluster Image Map were constructed to identify highly correlated joint clusters based on the composite PDUS scores. Based on the different approaches (best correlation, model optimization, etc.), representative joints were then selected from each group, which yielded several corresponding combinations of joints. Linear models were developed with these reduced joint sets as predictors of GLOESS, using data from 60% of patients randomly selected from the ULTIMATE study. The remaining 40% data were used for model validation and diagnostics.

Results

Five models were established with reduced pairs of joint sets (9–13 pairs). The joints included in each linear model are summarized in Table 1. All five models of reduced joint sets showed high correlation with GLOESS score of R2 ~ 0.95. Figure 1 depicts all the 5 models of reduced joint sets for PDUS-detected synovitis versus the actual GLOESS in secukinumab and placebo-secukinumab groups, with modified GLOESS scores using reduced PDUS joint sets demonstrating changes very close to that of validated GLOESS.

Table 1.

Joints included across five linear models, indicated by green shading

Joint pairsModel 1 (N=9)Model 2 (N=9)Model 3 (N=9)Model 4 (N=13)Model 5 (N=12)ElbowKneeMTP2WristMCP1DIP4MTP4MCP2MCP4MCP5PIP3PIP4PIP1, PIP5DIP2DIP3, DIP5MTP1MTP5ShoulderTibiotalar

N, number of joint pairs used in model. DIP, distal interphalangeal; MCP, metacarpophalangeal; MTP, metatarsophalangeal;

PIP, proximal interphalangeal

Figure 1.

Reduced set of joints synovitis score vs GLOESS score

Conclusion

All models of reduced joint sets for PDUS-detected synovitis predicted GLOESS well. The next steps will be to document responsiveness and ability to discriminate between active and placebo treatment.

References

[1]D’Agostino MA, et al. Rheumatology (Oxford) 2021;keab628.

Disclosure of Interests

Maria-Antonietta D’Agostino Speakers bureau: AbbVie, BMS, Celgene, Eli Lilly, Janssen, Novartis, Roche, Sanofi, and UCB, Consultant of: AbbVie, BMS, Celgene, Eli Lilly, Janssen, Novartis, Roche, Sanofi, and UCB, Maarten Boers Consultant of: BMS, GSK, Novartis, Pfizer, Consultant of: BMS, GSK, Novartis and Pfizer, Georg Schett Speakers bureau: AbbVie, BMS, Celgene, Janssen, Lilly, Novartis, Roche and UCB, Philip G Conaghan Speakers bureau: AbbVie, AstraZeneca, BMS, Eli Lilly, Galapagos, Gilead, Novartis and Pfizer, Consultant of: AbbVie, AstraZeneca, BMS, Eli Lilly, Galapagos, Gilead, Novartis and Pfizer, Esperanza Naredo Speakers bureau: AbbVie, BMS, Celgene GmbH, Janssen, Lilly, Novartis, Pfizer, Roche, UCB, Grant/research support from: Honoraria for clinical trials from Abbvie, BMS and Novartis; Research Grants from Lilly, Peter Mandl Speakers bureau: AbbVie, BMS, Celgene, Janssen, Lilly, MSD, Novartis, Roche and UCB., Grant/research support from: AbbVie, BMS, Celgene, Janssen, Lilly, MSD, Novartis, Roche and UCB., Philippe Carron Speakers bureau: AbbVie, Bristol Myers Squibb, Celgene Corporation, Eli Lilly, Gilead, Merck Sharp Dohme, Novartis, Pfizer, and UCB, Consultant of: AbbVie, Bristol Myers Squibb, Celgene Corporation, Eli Lilly, Gilead, Merck Sharp Dohme, Novartis, Pfizer, and UCB, Grant/research support from: Merck Sharp Dohme, Pfizer and UCB, Marina Backhaus Speakers bureau: BMS, Gilead, Jonsson, MSD, Novartis, Pfizer, Roche and UCB, Consultant of: BMS, Gilead, Jonsson, MSD, Novartis, Pfizer, Roche, UCB, Grant/research support from: BMS, Gilead, Jonsson, MSD, Novartis, Pfizer, Roche, UCB, Alejandra López-Rodríguez Speakers bureau: Eli Lilly, GSK, Janssen, Novartis, Roche and UCB, Consultant of: Eli Lilly, GSK, Janssen, Novartis, Roche and UCB, Petra Hanova: None declared, Punit Goyanka Employee of: Novartis, Braja Gopal Sahoo Employee of: Novartis, Corine Gaillez Shareholder of: Shareholder of Novartis and BMS, Employee of: Novartis, Weibin Bao Employee of: Novartis

POS1065 IMPACT OF HYPERURICEMIA ON CLINICAL PHENOTYPE, COMORBIDITIES, AND RESPONSE TO SECUKINUMAB IN PSORIATIC ARTHRITIS: POST HOC ANALYSIS OF FUTURE AND MAXIMISE STUDIES

Background

Hyperuricemia (HU) is a metabolic abnormality associated with psoriasis (PsO) and psoriatic arthritis (PsA)1. The prevalence of HU is 2–13% in general population, 19–20% in PsO patients (pts), and 27–32% in PsA pts1,2. Pts with PsO/PsA are at significantly increased risk of HU and development of gout1. The pathogenic role of chronic HU in the development and maintenance of PsA is based on epidemiological, clinical, and fundamental arguments and hence does not appear fortuitous. These processes can influence each other3. Moreover, PsA with HU has been shown to be more peripheral, destructive, and challenging to treat4.

Objectives

To evaluate the impact of HU on PsA in terms of clinical presentation, severity, comorbidities, and response to secukinumab (SEC) over 1-year.

Methods

This post hoc analysis included pooled data from PsA pts enrolled in the FUTURE 2–5 and MAXIMISE phase 3 trials. Pts were stratified into 2 groups based on baseline (BL) serum uric acid (SUA) level (HU: ≥360 µmol/L; without HU: <360 µmol/L and no history of gout and/or uric acid lowering therapies [ULT]). Demographic and disease characteristics, PsA and therapeutic history, and comorbidities data, were collected at BL. Evaluations included ACR20/50/70 responses, Psoriasis Area and Severity Index (PASI) 90 response, resolution of enthesitis and dactylitis, Health Assessment Questionnaire Disability Index (HAQ-DI), and mean change in SUA level, up to Week 52. All analyses were performed at a descriptive level and data presented as observed.

Results

Overall, 2504 PsA pts were included in the analysis, of which 822 (32.8%) had HU (62 [2.5%] with gout; 49 [2.0%] treated with ULT). At BL, pts with HU were mostly male (76.0% vs 34.2%) and had a higher body mass index (30.9 vs 28.3 kg/m2) with more comorbidities, such as hypertension (43.8% vs 31.3%), compared to pts without HU. A higher proportion of pts with HU had dactylitis (34.5% vs 25.9%), and PsO (48.3% vs 36.3%) with a greater mean PASI score (13.6 vs 10.2), compared to pts without HU (Table 1). The proportion of pts achieving ACR50, resolution of enthesitis/dactylitis, and mean change in HAQ-DI score were comparable up to Week 52 irrespective of BL HU status. The PASI90 response rate was higher in pts without HU with SEC 150 mg (with and without load) and similar in SEC 300 mg group irrespective of BL HU status (Figure 1).

Table 1.

Demographics and baseline characteristics

Parameters, mean ± SD unless specifiedWith hyperuricemia (N=822)Without hyperuricemia (N=1682)Age (Years)48.5 ± 12.4148.3 ± 12.19Gender (Male), n (%)625 (76.0)576 (34.2)Weight (kg)92.71 ± 18.6279.59 ± 17.55BMI (kg/m2)30.90 ± 5.8628.33 ± 5.91History of hypertension, n (%)360 (43.8)526 (31.3)History of diabetes mellitus, n (%)85 (10.3)144 (8.6)TJC20.6 ± 15.5221.3 ± 16.25SJC10.9 ± 9.3110.8 ± 9.13Enthesitis, n (%)412 (50.1)852 (50.7)Dactylitis, n (%)284 (34.5)436 (25.9)Evidence of current psoriasis; n (%)397 (48.3)611 (36.3)Mean PASI score*13.61 ± 11.0310.16 ± 9.13TNFi naїve, n (%)477 (58.0)938 (55.8)MTX use at randomization, n (%)321 (39.1)685 (40.7)Serum uric acid (µmol/L)420.7 ± 57.11274.9 ± 51.98CRP (mg/L)11.6 ± 18.6610.7 ± 23.36

*not collected in MAXMISE

BMI, body mass index; CRP, C-reactive protein; MTX, methotrexate; SJC, swollen joint count; TJC, tender joint count; TNFi, tumor necrosis factor inhibitor

Conclusion

In this pooled analysis of SEC PsA studies, pts with HU reported a higher prevalence of hypertension, with more clinical dactylitis, and more PsO, with higher PASI score compared to pts without HU. Efficacy across all musculoskeletal manifestations was similar with SEC 150 and 300 mg; while PASI90 response rate was slightly better in patients without HU with SEC 150 mg, and similar with SEC 300 mg irrespective of HU status, at 1-year.

References

[1]Tripolino C, et al. Front Med. 2021;8:737573

[2]AlJohani R, et al. J Rheumatol. 2018;45(2):213–7

[3]Felten R, et al. Clin Rheumatol. 2020;39:1405–13

[4]Widawski L, et al. Clin Rheumatol. 2022. https://doi.org/10.1007/s10067-022-06061-x

Disclosure of Interests

Renaud FELTEN Consultant of: Novartis (Advisory board), Laura Widawski: None declared, Lionel Spielmann: None declared, Corine Gaillez Shareholder of: Novartis, Employee of: Novartis, Weibin Bao Shareholder of: Novartis, Employee of: Novartis, Jacques-Eric Gottenberg Consultant of: Novartis (Advisory board), Pierre-Marie Duret: None declared, Laurent Messer: None declared

Gene Regulatory Identification Based on the Novel Hybrid Time-Delayed Method

Gene regulatory network (GRN) inference with biology data is a difficult and serious issue in the field of system biology. In order to detect the direct associations of GRN more accurately, a novel two-step GRN inference technique based on the time-delayed correlation coefficient (TDCC) and time-delayed complex-valued S-system model (TDCVSS) is proposed. First, a TDCC algorithm is utilized to construct an initial network. Second, a TDCVSS model is utilized to prune the network topology in order to delete false-positive regulatory relationships for each target gene. The complex-valued restricted additive tree and complex-valued differential evolution are proposed to approximate the optimal TDCVSS model. Finally, the overall network could be inferred by integrating the regulations of all target genes. Two real gene expression datasets from E. coli and S. cerevisiae gene networks are utilized to evaluate the performances of our proposed two-step GRN inference algorithm. The results demonstrated that the proposed algorithm could infer GRN more correct than classical methods and time-delayed methods.

[Pituicytoma: a clinicopathological analysis of twenty-one cases]

Objective: To investigate the clinicopathological features and treatment strategies of pituicytoma. Methods: Twenty-one cases of pituicytoma were collected at the First Affiliated Hospital of Nanjing Medical University and Jinling Hospital, Nanjing, China from 2009 to 2020. The clinical data of 21 pituicytoma patients was retrospectively analyzed, and the relevant literature was reviewed. Results: Twenty-one patients aged 4 to 68 years, including 8 males and 13 females. All patients underwent surgical treatment. Histologically, the tumor was consisted almost entirely of elongate, bipolar spindle cells arranged in a fascicular or storiform pattern. Mitotic figures were rare. Immunohistochemically, tumor cells were diffusely positive for S-100 protein (21/21), vimentin (15/15) and TTF1 (14/14), while they were weakly or focally positive for GFAP (13/16) and EMA (6/12). CKpan was negative in all cases and Ki-67 proliferation index was low (<5%). Among the 18 patients with follow-up, all survived and 2 relapsed after surgery. Conclusions: Pituicytoma is a rare low-grade glioma of the sellar area. It is easily confused with other sellar tumors. Preoperative diagnosis is difficult. It needs to be confirmed by histopathology and immunohistochemistry. Microsurgery is the main treatment method at present.

Active disease-related compound identification based on capsule network

Pneumonia, especially corona virus disease 2019 (COVID-19), can lead to serious acute lung injury, acute respiratory distress syndrome, multiple organ failure and even death. Thus it is an urgent task for developing high-efficiency, low-toxicity and targeted drugs according to pathogenesis of coronavirus. In this paper, a novel disease-related compound identification model-based capsule network (CapsNet) is proposed. According to pneumonia-related keywords, the prescriptions and active components related to the pharmacological mechanism of disease are collected and extracted in order to construct training set. The features of each component are extracted as the input layer of capsule network. CapsNet is trained and utilized to identify the pneumonia-related compounds in Qingre Jiedu injection. The experiment results show that CapsNet can identify disease-related compounds more accurately than SVM, RF, gcForest and forgeNet.

Hypertension-Related Drug Activity Identification Based on Novel Ensemble Method

Hypertension is a chronic disease and major risk factor for cardiovascular and cerebrovascular diseases that often leads to damage to target organs. The prevention and treatment of hypertension is crucially important for human health. In this paper, a novel ensemble method based on a flexible neural tree (FNT) is proposed to identify hypertension-related active compounds. In the ensemble method, the base classifiers are Multi-Grained Cascade Forest (gcForest), support vector machines (SVM), random forest (RF), AdaBoost, decision tree (DT), Gradient Boosting Decision Tree (GBDT), KNN, logical regression, and naïve Bayes (NB). The classification results of nine classifiers are utilized as the input vector of FNT, which is utilized as a nonlinear ensemble method to identify hypertension-related drug compounds. The experiment data are extracted from hypertension-unrelated and hypertension-related compounds collected from the up-to-date literature. The results reveal that our proposed ensemble method performs better than other single classifiers in terms of ROC curve, AUC, TPR, FRP, Precision, Specificity, and F1. Our proposed method is also compared with the averaged and voting ensemble methods. The results reveal that our method could identify hypertension-related compounds more accurately than two classical ensemble methods.

Reverse engineering gene regulatory network based on complex-valued ordinary differential equation model

BACKGROUND

The growing researches of molecular biology reveal that complex life phenomena have the ability to demonstrating various types of interactions in the level of genomics. To establish the interactions between genes or proteins and understand the intrinsic mechanisms of biological systems have become an urgent need and study hotspot.

RESULTS

In order to forecast gene expression data and identify more accurate gene regulatory network, complex-valued version of ordinary differential equation (CVODE) is proposed in this paper. In order to optimize CVODE model, a complex-valued hybrid evolutionary method based on Grammar-guided genetic programming and complex-valued firefly algorithm is presented.

CONCLUSIONS

When tested on three real gene expression datasets from E. coli and Human Cell, the experiment results suggest that CVODE model could improve 20-50% prediction accuracy of gene expression data, which could also infer more true-positive regulatory relationships and less false-positive regulations than ordinary differential equation.

Efficient framework for predicting MiRNA-disease associations based on improved hybrid collaborative filtering

BACKGROUND

Accumulating studies indicates that microRNAs (miRNAs) play vital roles in the process of development and progression of many human complex diseases. However, traditional biochemical experimental methods for identifying disease-related miRNAs cost large amount of time, manpower, material and financial resources.

METHODS

In this study, we developed a framework named hybrid collaborative filtering for miRNA-disease association prediction (HCFMDA) by integrating heterogeneous data, e.g., miRNA functional similarity, disease semantic similarity, known miRNA-disease association networks, and Gaussian kernel similarity of miRNAs and diseases. To capture the intrinsic interaction patterns embedded in the sparse association matrix, we prioritized the predictive score by fusing three types of information: similar disease associations, similar miRNA associations, and similar disease-miRNA associations. Meanwhile, singular value decomposition was adopted to reduce the impact of noise and accelerate predictive speed.

RESULTS

We then validated HCFMDA with leave-one-out cross-validation (LOOCV) and two types of case studies. In the LOOCV, we achieved 0.8379 of AUC (area under the curve). To evaluate the performance of HCFMDA on real diseases, we further implemented the first type of case validation over three important human diseases: Colon Neoplasms, Esophageal Neoplasms and Prostate Neoplasms. As a result, 44, 46 and 44 out of the top 50 predicted disease-related miRNAs were confirmed by experimental evidence. Moreover, the second type of case validation on Breast Neoplasms indicates that HCFMDA could also be applied to predict potential miRNAs towards those diseases without any known associated miRNA.

CONCLUSIONS

The satisfactory prediction performance demonstrates that our model could serve as a reliable tool to guide the following research for identifying candidate miRNAs associated with human diseases.

POS1051 SECUKINUMAB IMPROVES PHYSICAL FUNCTION AND INHIBITS STRUCTURAL DAMAGE IN PsA PATIENTS WITH SUSTAINED REMISSION OR LOW DISEASE ACTIVITY: RESULTS FROM A PHASE 3 STUDY

Background:

Disease Activity Index for Psoriatic Arthritis (DAPSA) or the minimal disease activity (MDA) are considered for defining remission (REM) or low disease activity (LDA) in secukinumab (SEC) treated patients (pts) with PsA (Psoriatic Arthritis).1 Currently, limited SEC data are available on pts with PsA achieving sustained REM in clinical trials or real-world evidence, using these stringent criteria.

Objectives:

To report an exploratory analysis on achievement of sustained REM/LDA in pts with PsA treated with SEC and impact on structural outcomes, physical function and health-related quality of life (HRQoL) in the FUTURE 5 study (NCT02404350).

Methods:

FUTURE 5 is a randomised, double-blind, placebo-controlled 2-year phase 3 trial in pts with active PsA.2 Pts randomised to SEC 150 mg could be escalated to 300 mg from Week (Wk) 52 to 104, based on investigators’ judgement. The pts were categorised as either not achieving REM/LDA, achieving it once only or sustained REM/LDA, which was defined as pts who achieved REM/LDA between Wks 24-52 and maintained the same response at least 2 of the next 6 visits (visit every 8 Wks). Of pts who did not achieve REM/LDA, achieved REM/LDA (VLDA, DAPSA REM, MDA, DAPSA LDA+REM) between Wk 24 and 52, the relationship between absence of REM/LDA, REM/LDA, sustained REM/LDA, proportion of pts with non-radiographic progression (assessed using the van der Heijde [mTSS]), physical function (health assessment questionnaire disability index [HAQ-DI]), and short form-36 physical component score [SF-36 PCS])3 were assessed.

Results:

In total, 996 pts were randomised to one of 4 treatment groups: SEC 300 mg loading dose (LD; N=222), SEC 150 mg LD (N=220), SEC 150 mg no loading dose (NL; N=222), and placebo (N=332). The baseline clinical characteristics were comparable across treatment groups. Majority of pts could achieve either sustained MDA/sustained DAPSA LDA+REM (Figure 1). Pts achieving REM/LDA, whether at one visit or consistently, showed improved physical function and SF36-PCS at Wk 104.3A high proportion of pts did not show radiographic progression at Wk 104 irrespective of achievement of REM/LDA category (Table 1).

Conclusion:

The majority of patients treated with secukinumab were able to achieve sustained LDA. Sustained LDA/REM was associated with improved HRQoL, physical function and inhibition of structural damage progression.

References:

[1]Coates LC, et al. J Rheumatol. 2018;46(1):38–42.

[2]Van der Heijde D, et al. Rheumatology. 2020;59(6):1325–1334.

[3]Coates L, et al. [0353]. Arthritis Rheumatol. 2020;72 (suppl 10).

Figure 1.

Proportion of patients achieving VLDA/MDA/DAPSA REM/DAPSA REM+LDASus-tained REM/LDA was defined if the same response was achieved at least 2 out of the next 6 visits (every 8 weeks), respectively. DAPSA, Disease activity in Psoriatic Arthritis; LD, loading dose; LDA, Low Disease Activity; MDA, Minimal Disease Activity; N, number of randomised patients assessed at both Week 24 and 104; NL, without loading dose; REM, remission; SEC, secukinumab; VLDA, Very Low Disease Activity

Table 1.

Percentage of vdH-mTSS no progression at Week 104 (change from baseline ≤0.5) by REM/LDA and sustained REM/LDA status

REM and LDA composite indices, n (%)Treatment groupNo REM/LDAREM/LDA only onceSustained REM/LDAMDASEC 150 mg LD64 (75.3)16 (80.0)76 (86.4)SEC 150 mg NL56 (75.7)15 (78.9)69 (82.1)SEC 300 mg LD58 (79.5)19 (95.0)100 (94.3)VLDASEC 150 mg LD108 (78.8)15 (83.3)30 (88.2)SEC 150 mg NL95 (75.4)13 (81.3)32 (91.4)SEC 300 mg LD 115 (84.6)17 (94.4)45 (100.0)DAPSA REMSEC 150 mg LD77 (76.2)11 (78.6)46 (92.0)SEC 150 mg NL65 (71.4)10 (76.9)50 (87.7)SEC 300 mg LD82 (83.7)14 (93.3)63 (96.9)DAPSA LDA + REMSEC 150 mg LD29 (70.7)16 (84.2)80 (84.2)SEC 150 mg NL23 (71.9)15 (75.0)79 (79.0)SEC 300 mg LD39 (88.6)21 (84.0)97 (89.0)

Sustained REM/LDA was defined if the same response was achieved at least twice out of the next 6 visits (every 8 weeks), respectively. n, number of evaluable patients; vdH-mTSS, van der Heijde- modified total Sharp score

Disclosure of Interests:

Laura C Coates Consultant of: Abbvie, Amgen, Biogen, Boehringer Ingelheim, Celgene, Galapagos, Gilead, Janssen, Lilly, Novartis, Pfizer and UCB, Grant/research support from: Abbvie, Celgene, Lilly, Novartis and Pfizer, Philip J Mease Speakers bureau: AbbVie, Amgen, Janssen, Lilly, Novartis, Pfizer and UCB, Consultant of: AbbVie, Amgen, BMS, Boehringer Ingelheim, Galapagos, Celgene, Genentech, Gilead, Janssen, Lilly, Novartis, Pfizer, SUN Pharma, and UCB, Grant/research support from: AbbVie, Amgen, BMS, Celgene, Galapagos, Gilead, Janssen, Lilly, Novartis, Pfizer, SUN, and UCB, Dafna D Gladman Consultant of: Abbvie, Amgen, BMS, Celgene, Eli Lilly, Gilead, Galapagos, Janssen, Novartis, Pfizer and UCB, Grant/research support from: Abbvie, Amgen, BMS, Celgene, Eli Lilly, Gilead, Galapagos, Janssen, Novartis, Pfizer and UCB, Sandra Navarra Speakers bureau: Pfizer, Novartis, Astra-Zeneca, Janssen, Lilly, and Astellas, Consultant of: Pfizer, Novartis, Astra-Zeneca, Janssen, Lilly, and Astellas, Weibin Bao Shareholder of: Novartis, Employee of: Novartis, Corine Gaillez Shareholder of: Novartis and BMS, Employee of: Novartis.

POS1044 EFFECT OF SECUKINUMAB VERSUS ADALIMUMAB ON ACR CORE COMPONENTS AND HEALTH-RELATED QUALITY OF LIFE IN PATIENTS WITH PSORIATIC ARTHRITIS: RESULTS FROM THE EXCEED STUDY

Background:

EXCEED (NCT02745080) was the first fully blinded head-to-head trial to evaluate the efficacy and safety of secukinumab (SEC) versus (vs) adalimumab (ADA) monotherapy in patients with active psoriatic arthritis (PsA) with a primary endpoint of American College of Rheumatology (ACR) 20 at Week 52. Although SEC narrowly missed statistical significance for superiority vs ADA, numerically higher response for other musculoskeletal endpoints and composite indices were observed with SEC.1

Objectives:

To explore the effect of SEC and ADA on ACR core components, function and Health-related Quality of Life (HRQoL) outcomes.

Methods:

Patients were randomised 1:1 to receive SEC 300 mg (N=426) subcutaneous (s.c.) at baseline, Week 1-4, followed by every 4 weeks until Week 48 or ADA 40 mg (N=427) s.c. at baseline followed by same dosing every 2 weeks until Week 50. The primary, key secondary and some exploratory endpoints at Week 52 were previously reported.1 A supportive analysis for ACR50 response using logistic regression model and trimmed means model for Health Assessment Questionnaire-Disability Index (HAQ-DI) with gender and smoking status as factors was performed to adjust for imbalances in baseline characteristics. An exploratory analysis of ACR core components with SEC vs ADA at Week 52 was conducted using a mixed-effects repeated measures model that included tender and swollen joint counts, patient and physician global assessment, PsA pain (VAS) and erythrocyte sedimentation rate. HRQoL variables were also exploratory and assessed based on Short Form Health Survey Physical/Mental Component Summary (SF-36 PCS/MCS) scores and Dermatology Life Quality Index (DLQI).

Results:

The demographic and baseline disease characteristics were comparable across treatment groups, except for an imbalance in sex (females: 51.2% vs 46.4%) and smoking status (yes: 21.8% vs 17.8%) in SEC and ADA group, respectively. At Week 52, ACR50 responses were 49.0% and 44.8% (P=0.0929) and HAQ-DI mean change from baseline were −0.69 and −0.58 (P=0.0314) in SEC and ADA treatment groups, respectively after adjusting for gender and smoking status. No major difference across ACR core components was observed in both treatment groups at Week 52 (Table 1). At Week 52, SEC presented similar improvement in SF-36 PCS/MCS score and numerically higher improvement in DLQI compared to ADA (Figure 1).

Conclusion:

Secukinumab provided similar improvements in ACR core components and SF-36 based quality of life at Week 52 with adalimumab. Greater improvement in HAQ-DI response and DLQI was demonstrated with secukinumab compared to adalimumab.

References:

[1]McInnes IB, et al. Lancet. 2020; 395:1496–505.

Table 1.

ACR Core Components at Week 52

VariablesSecukinumab 300 mg(N=426)Adalimumab 40 mg(N=427)P-valueBL, mean ± SELSM change from BL ± SEBL, mean ± SELSM change from BL ± SETender joint score(based on 78 joints)19.4 ± 13.86−14.27 ± 0.4420.6 ± 14.81−13.90 ± 0.450.5549Swollen joint score(based on 76 joints)9.7 ± 7.30−8.41 ± 0.1910.2 ± 7.86−8.06 ± 0.200.1962Patients global assessment64.0 ± 19.67−33.81 ± 1.1461.9 ± 20.75−31.61 ± 1.190.1825Physicians global assessment60.0 ± 17.12−46.24 ± 0.8061.4 ± 15.92−43.63 ± 0.840.0243Psoriatic arthritis pain (VAS)58.6 ± 23.49−30.21 ± 1.1857.9 ± 22.42−29.44 ± 1.230.6500Erythrocyte sedimentation rate (mm/h)23.8 ± 18.93−9.63 ± 0.6223.9 ± 17.99−9.28 ± 0.640.7029

LS mean and nominal P-values are from a mixed-effects repeated measures model with treatment group, analysis visit as factors, weight and BL score as covariates, and by treatment and BL score as interaction terms, unstructured covariance structure. ACR, American College of Rheumatology; BL, baseline; LSM, least squares mean; N, total number of randomised patients; SE, standard error; VAS, visual analogue scale

Figure 1.

HRQoL Analysis at Week 52

Disclosure of Interests:

Philippe Goupille Speakers bureau: AbbVie, Amgen, Biogen, BMS, Celgene, Chugai, Janssen, Eli Lilly, Medac, MSD, Nordic Pharma, Novartis, Pfizer, Sanofi and UCB, Consultant of: AbbVie, Amgen, Biogen, BMS, Celgene, Chugai, Janssen, Eli Lilly, Medac, MSD, Nordic Pharma, Novartis, Pfizer, Sanofi and UCB, Grant/research support from: AbbVie, Amgen, Biogen, BMS, Celgene, Chugai, Janssen, Eli Lilly, Medac, MSD, Nordic Pharma, Novartis, Pfizer, Sanofi and UCB, Frank Behrens Paid instructor for: Eli Lilly, Consultant of: Pfizer, AbbVie, Sanofi, Eli Lilly, Novartis, Genzyme, Boehringer Ingelheim, Janssen, MSD, Celgene, Roche and Chugai, Grant/research support from: Pfizer, Janssen, Chugai, Celgene and Roche, Laura C Coates Consultant of: AbbVie, Amgen, Boehringer Ingelheim, Biogen, BMS, Celgene, Domain, Eli Lilly, Gilead, GSK, Janssen, Medac, Novartis, Pfizer, Serac and UCB, Grant/research support from: AbbVie, Amgen, Celgene, Eli Lilly, Janssen, Novartis, Pfizer and UCB, Jordi Gratacos-Masmitja Speakers bureau: AbbVie, Amgen, BMS, Celgene, Janssen, Eli Lilly, Novartis and Pfizer, Consultant of: AbbVie, Amgen, BMS, Celgene, Janssen, Eli Lilly, Novartis and Pfizer, Grant/research support from: AbbVie, Amgen, BMS, Celgene, Janssen, Eli Lilly, Novartis and Pfizer, Philip J Mease Speakers bureau: AbbVie, Amgen, Genentech, Janssen, Eli Lilly, Merck, Novartis, Pfizer, and UCB, Consultant of: AbbVie, Amgen, Bristol-Myers Squibb, Boehringer Ingelheim, Galapagos, Celgene, Genentech, Gilead, Janssen, Eli Lilly, Novartis, Pfizer, SUN Pharma, and UCB, Grant/research support from: AbbVie, Amgen, Bristol-Myers Squibb, Celgene, Galapagos, Genentech, Gilead, Janssen, Eli Lilly, Merck, Novartis, Pfizer, SUN Pharma, and UCB, Dafna D Gladman Consultant of: Amgen, AbbVie, BMS, Celgene, Eli Lilly, Gilead, Galapagos, Janssen, Novartis, Pfizer and UCB, Grant/research support from: Amgen, AbbVie, Celgene, Eli Lilly, Janssen, Novartis, Pfizer and UCB, Peter Nash Speakers bureau: Novartis, Abbvie, Roche, Pfizer, BMS, Janssen, Celgene, UCB, Eli Lilly, MSD, Sanofi, Gilead, Consultant of: Novartis, Abbvie, Roche, Pfizer, BMS, Janssen, Celgene, UCB, Eli Lilly, MSD, Sanofi, Gilead, Grant/research support from: Novartis, Abbvie, Roche, Pfizer, BMS, Janssen, Celgene, UCB, Eli Lilly, MSD, Sanofi, Gilead, Arthur Kavanaugh Consultant of: AbbVie, Amgen, Celgene, Eli Lilly, Janssen, Novartis, and UCB, Grant/research support from: AbbVie, Amgen, Celgene, Eli Lilly, Janssen, Novartis, and UCB, Ruvie Martin Shareholder of: Novartis, Employee of: Novartis, Weibin Bao Shareholder of: Novartis, Employee of: Novartis, Corine Gaillez Shareholder of: Novartis and BMS, Employee of: Novartis, Iain McInnes Speakers bureau: AbbVie, Amgen, Bristol-Myers Squibb, Celgene, Janssen, Eli Lilly, Novartis, Pfizer, and UCB, Consultant of: AbbVie, Amgen, Bristol-Myers Squibb, Celgene, Janssen, Eli Lilly, Novartis, Pfizer, and UCB, Grant/research support from: AbbVie, Amgen, Bristol-Myers Squibb, Celgene, Janssen, Eli Lilly, Novartis, Pfizer, and UCB.

Efficacy of secukinumab and adalimumab in patients with psoriatic arthritis and concomitant moderate-to-severe plaque psoriasis: results from EXCEED, a randomized, double-blind head-to-head monotherapy study

Background

Secukinumab [an interleukin (IL)‐17A inhibitor] has demonstrated significantly higher efficacy vs. etanercept (a tumour necrosis factor inhibitor) and ustekinumab (an IL‐12/23 inhibitor) in patients with moderate‐to‐severe plaque psoriasis.

Objectives

To report 52‐week results from a prespecified analysis of patients with active psoriatic arthritis (PsA) having concomitant moderate‐to‐severe plaque psoriasis from the head‐to‐head EXCEED monotherapy study comparing secukinumab with adalimumab.

Methods

Patients were randomized to receive secukinumab 300 mg via subcutaneous injection at baseline, week 1–4, and then every 4 weeks until week 48 or adalimumab 40 mg via subcutaneous injection every 2 weeks from baseline until week 50. Assessments in patients with concomitant moderate‐to‐severe psoriasis, defined as having affected body surface area > 10% or Psoriasis Area and Severity Index (PASI) ≥ 10 at baseline, included musculoskeletal, skin and quality‐of‐life outcomes. Missing data were handled using multiple imputation.

Results

Of the 853 patients [secukinumab (N = 426), adalimumab (N = 427)], 211 (24·7%) had concomitant moderate‐to‐severe psoriasis [secukinumab (N = 110, 25·8%), adalimumab (N = 101, 23·7%)]. Up to week 50, 5·5% of patients discontinued secukinumab vs.17·8% in the adalimumab group. The proportion of patients who achieved American College of Rheumatology (ACR) 20 response was 76·4% with secukinumab vs. 68·3% with adalimumab (P = 0·175), PASI 100 response was 39·1% vs. 23·8% (P = 0·013), and simultaneous improvement in ACR 50 and PASI 100 response at week 52 was 28·2% vs. 17·7%, respectively (P = 0·06). Secukinumab demonstrated consistently higher responses vs. adalimumab across skin endpoints.

Conclusions

This prespecified analysis in PsA patients with concomitant moderate‐to‐severe plaque psoriasis in the EXCEED study provides further evidence that IL‐17 inhibitors offer a comprehensive biological treatment to manage the concomitant features of psoriasis and PsA.

What is already known about this topic?

Secukinumab, an interleukin‐17A inhibitor, has previously been reported to have significantly higher efficacy in head‐to‐head trials vs. etanercept and ustekinumab in patients with moderate‐to‐severe plaque psoriasis.

What does this study add?

The results of the study provide valuable head‐to‐head data on the efficacy of two biologics with different mechanisms of action (secukinumab and adalimumab) as first‐line biological monotherapy for patients with psoriatic arthritis and concomitant moderate‐to‐severe plaque psoriasis.

The findings of this study can further help physicians to make informed and evidence‐based decisions for the treatment of patients with active psoriatic arthritis who have concomitant moderate‐to‐severe plaque psoriasis.

Linked Comment: E. Sbidian and L. Pina‐Vegas. Br J Dermatol 2021; 185:1085.

The risk of malignancy in patients with secukinumab-treated psoriasis, psoriatic arthritis and ankylosing spondylitis: analysis of clinical trial and postmarketing surveillance data with up to five years of follow-up

BACKGROUND

Data on the use of biologic therapy and malignancy risk are inconsistent due to limited long-term robust studies.

OBJECTIVES

To assess the malignancy risk in patients with secukinumab-treated psoriasis, psoriatic arthritis (PsA) and ankylosing spondylitis (AS).

METHODS

This integrated safety analysis from both the secukinumab clinical trial programme and postmarketing safety surveillance data included any patient receiving at least one approved dose of secukinumab with a maximum of 5 years of follow-up. Safety analyses evaluated the rate of malignancy using exposure-adjusted incidence rates [EAIR; incidence rates per 100 patient treatment-years (PTY)]. Standardized incidence ratios (SIRs) were reported using the Surveillance, Epidemiology, and End Results Program (SEER) database as a reference population. Crude incidence of malignancy was also reported using postmarketing surveillance data.

RESULTS

Safety data from 49 clinical trials with secukinumab-treated patients were included: 10 685 patients with psoriasis, 2523 with PsA and 1311 with AS. Across indications over a 5-year period, the EAIR of malignancy was 0·85 per 100 PTY [95% confidence interval (CI) 0·74-0·98] in secukinumab-treated patients, corresponding to 204 patients per 23 908 PTY. Overall, the observed vs. expected number of malignancies from secukinumab clinical trial data were comparable, as indicated by an SIR of 0·99 (95% CI 0·82-1·19) across indications. The estimated crude cumulative incidence reporting rate per 100 PTY for malignancy was 0·27 in the postmarketing surveillance data across indications with a cumulative exposure of 285 811 PTY.

CONCLUSIONS

In this large safety analysis, the risk of malignancy was low for up to 5 years of secukinumab treatment. These data support the long-term use of secukinumab in these indications.

A graph auto-encoder model for miRNA-disease associations prediction

Emerging evidence indicates that the abnormal expression of miRNAs involves in the evolution and progression of various human complex diseases. Identifying disease-related miRNAs as new biomarkers can promote the development of disease pathology and clinical medicine. However, designing biological experiments to validate disease-related miRNAs is usually time-consuming and expensive. Therefore, it is urgent to design effective computational methods for predicting potential miRNA-disease associations. Inspired by the great progress of graph neural networks in link prediction, we propose a novel graph auto-encoder model, named GAEMDA, to identify the potential miRNA-disease associations in an end-to-end manner. More specifically, the GAEMDA model applies a graph neural networks-based encoder, which contains aggregator function and multi-layer perceptron for aggregating nodes' neighborhood information, to generate the low-dimensional embeddings of miRNA and disease nodes and realize the effective fusion of heterogeneous information. Then, the embeddings of miRNA and disease nodes are fed into a bilinear decoder to identify the potential links between miRNA and disease nodes. The experimental results indicate that GAEMDA achieves the average area under the curve of $93.56\pm 0.44\%$ under 5-fold cross-validation. Besides, we further carried out case studies on colon neoplasms, esophageal neoplasms and kidney neoplasms. As a result, 48 of the top 50 predicted miRNAs associated with these diseases are confirmed by the database of differentially expressed miRNAs in human cancers and microRNA deregulation in human disease database, respectively. The satisfactory prediction performance suggests that GAEMDA model could serve as a reliable tool to guide the following researches on the regulatory role of miRNAs. Besides, the source codes are available at https://github.com/chimianbuhetang/GAEMDA.

FRI0272 SECUKINUMAB DEMONSTRATES A CONSISTENT SAFETY PROFILE IN PATIENTS WITH PSORIASIS, PSORIATIC ARTHRITIS AND ANKYLOSING SPONDYLITIS OVER LONG TERM: UPDATED POOLED SAFETY ANALYSES

Background:

Pooled safety data has been reported with secukinumab (SEC) in patients (pts) with Psoriatic arthritis (PsA), Ankylosing Spondylitis (AS) and Psoriasis (PsO).1

Objectives:

To report longer-term safety data of SEC treatment in PsA, AS, PsO pts up to 5 years.

Methods:

The integrated clinical trial safety dataset included data pooled from 28 randomised controlled clinical trials of SEC 300 or 150 or 75 mg in PsO (11 Phase 3 and 8 Phase 4 trials), PsA (5 Phase 3 trials), and AS (4 Phase 3 trials), along with post-marketing safety surveillance data with a cut-off date of 25 December 2018. Adverse events (AEs) were reported as exposure-adjusted incident rates (EAIRs) per 100 pt-years. Analyses included all pts who received ≥1 dose of SEC.

Results:

A total of 12637 pts (8819, 2678 and 1140 pts with PsO, PsA and AS, with an exposure of 15063.1, 5984.6 and 3527.2 pt-years, respectively) were included. The most frequent AE was upper respiratory tract infection and EAIR per 100 pt-years for IBD, malignancies and MACE remained low. The EAIR per 100 pt-years for adverse events (AEs) of special interest are reported in Table 1. The cumulative post-marketing exposure to SEC was estimated to be ~285,811 pt-years across the approved indications. Safety data from post-marketing surveillance are reported in Table 2.

Table 1.

Selected AEs of interest with SEC across pooled clinical trials

VariablePsOPsAASSECN=8819SECN=2678SECN=1140Exposure (Days), Mean (SD)623.9 (567.7)816.2 (580.7)1130.1 (583.0)Death, n (%)15 (0.2)13 (0.5)10 (0.9)Selected AE’s of interest, EAIR (95% CI)Serious infections11.4 (1.2, 1.6)1.8 (1.5, 2.2)1.2 (0.9, 1.6)Candidainfections22.9 (2.7, 3.2)1.5 (1.2, 1.9)0.7 (0.5, 1.1)IBD3Crohn’s disease3Ulcerative colitis30.01 (0.0, 0.05)0.1 (0.05, 0.2)0.1 (0.08, 0.2)0.03 (0.0, 0.1)0.1 (0.04, 0.2)0.1 (0.04, 0.2)0.03 (0.0, 0.2)0.4 (0.24, 0.7)0.2 (0.1, 0.5)MACE40.4 (0.31, 0.5)0.4 (0.3, 0.6)0.7 (0.4, 1.0)Uveitis30.01 (0.0, 0.05)0.1 (0.04, 0.2)1.2 (0.9, 1.7)Malignancy50.9 (0.7, 1.0)1.0 (0.77, 1.3)0.5 (0.3, 0.8)

1Rates for system organ class;2Rates for high level term;3Rates for preferred term (PT; IBD for unspecified IBD);4Rates for Novartis MedDRA Query term;5Rates for standardized MedDRA query term – ‘malignancies and unspecified tumour’; EAIR, exposure adjusted incidence rate per 100 pt-years; N, number of pts in the analysis

Table 2.

Summary of SEC post-marketing safety

Exposure (PTY)PSUR126Dec14 -25Jun15PSUR226 Jun - 25Dec15PSUR326Dec15 -25Jun16PSUR426Jun -25Dec16PSUR526Dec16 -25Dec17PSUR626Dec17 -25Dec18Cumulative18387450168712854993744137325285811                    n (Reporting rate PTY)Serious infections89 (4.8)149 (2.0)232 (1.4)475 (1.7)649 (0.7)1841 (1.3)3980 (1.4)Malignancy2 (0.1)15 (0.2)21 (0.1)50 (0.2)225 (0.2)422 (0.3)788 (0.3)Total IBD4 (0.2)12 (0.2)37(0.2)46 (0.2)185 (0.2)340 (0.3)693 (0.2)MACE6 (0.3)15 (0.2)16 (0.1)39 (0.1)151 (0.2)238 (0.2)504 (0.2)

PSUR, periodic safety update report; PTY, pt-treatment years

Conclusion:

In this long-term analysis across clinical trials and post-marketing surveillance, of pts with PsO, PsA and AS, SEC was well tolerated, with a safety profile consistent with previous reports.1

Reference:

[1]Deodhar et al. Arthritis Research & Therapy (2019) 21:111.

Disclosure of Interests:

Atul Deodhar Grant/research support from: AbbVie, Eli Lilly, GSK, Novartis, Pfizer, UCB, Consultant of: AbbVie, Amgen, Boehringer Ingelheim, Bristol Myer Squibb (BMS), Eli Lilly, GSK, Janssen, Novartis, Pfizer, UCB, Speakers bureau: AbbVie, Amgen, Boehringer Ingelheim, Bristol Myer Squibb (BMS), Eli Lilly, GSK, Janssen, Novartis, Pfizer, UCB, Iain McInnes Grant/research support from: Bristol-Myers Squibb, Celgene, Eli Lilly and Company, Janssen, and UCB, Consultant of: AbbVie, Bristol-Myers Squibb, Celgene, Eli Lilly and Company, Gilead, Janssen, Novartis, Pfizer, and UCB, Xenofon Baraliakos Grant/research support from: Grant/research support from: AbbVie, BMS, Celgene, Chugai, Merck, Novartis, Pfizer, UCB and Werfen, Consultant of: AbbVie, BMS, Celgene, Chugai, Merck, Novartis, Pfizer, UCB and Werfen, Speakers bureau: AbbVie, BMS, Celgene, Chugai, Merck, Novartis, Pfizer, UCB and Werfen, Kristian Reich Grant/research support from: Affibody; Almirall; Amgen; Biogen; Boehringer Ingelheim; Celgene; Centocor; Covagen; Eli Lilly; Forward Pharma; Fresenius Medical Care; GlaxoSmithKline; Janssen; Kyowa Kirin; LEO Pharma; Medac; Merck; Novartis; Miltenyi Biotec; Ocean Pharma; Pfizer; Regeneron; Samsung Bioepis; Sanofi Genzyme; Takeda; UCB; Valeant and Xenoport., Consultant of: Affibody; Almirall; Amgen; Biogen; Boehringer Ingelheim; Celgene; Centocor; Covagen; Eli Lilly; Forward Pharma; Fresenius Medical Care; GlaxoSmithKline; Janssen; Kyowa Kirin; LEO Pharma; Medac; Merck; Novartis; Miltenyi Biotec; Ocean Pharma; Pfizer; Regeneron; Samsung Bioepis; Sanofi Genzyme; Takeda; UCB; Valeant and Xenoport., Speakers bureau: Affibody; Almirall; Amgen; Biogen; Boehringer Ingelheim; Celgene; Centocor; Covagen; Eli Lilly; Forward Pharma; Fresenius Medical Care; GlaxoSmithKline; Janssen; Kyowa Kirin; LEO Pharma; Medac; Merck; Novartis; Miltenyi Biotec; Ocean Pharma; Pfizer; Regeneron; Samsung Bioepis; Sanofi Genzyme; Takeda; UCB; Valeant and Xenoport., Alice B Gottlieb Grant/research support from:: Research grants, consultation fees, or speaker honoraria for lectures from: Pfizer, AbbVie, BMS, Lilly, MSD, Novartis, Roche, Sanofi, Sandoz, Nordic, Celltrion and UCB., Consultant of:: Research grants, consultation fees, or speaker honoraria for lectures from: Pfizer, AbbVie, BMS, Lilly, MSD, Novartis, Roche, Sanofi, Sandoz, Nordic, Celltrion and UCB., Speakers bureau:: Research grants, consultation fees, or speaker honoraria for lectures from: Pfizer, AbbVie, BMS, Lilly, MSD, Novartis, Roche, Sanofi, Sandoz, Nordic, Celltrion and UCB., Mark Lebwohl Grant/research support from: AbbVie, Amgen, Arcutis, AstraZeneca, Boehringer Ingelheim, Celgene, Clinuvel, Eli Lilly, Incyte, Janssen Research & Development, LLC, Kadmon Corp., LLC, Leo Pharmaceutucals, Medimmune, Novartis, Ortho Dermatologics, Pfizer, Sciderm, UCB, Inc., and ViDac, Consultant of: Allergan, Almirall, Arcutis, Inc., Avotres Therapeutics, BirchBioMed Inc., Boehringer-Ingelheim, Bristol-Myers Squibb, Cara Therapeutics, Castle Biosciences, Corrona, Dermavant Sciences, Evelo, Foundation for Research and Education in Dermatology, Inozyme Pharma, LEO Pharma, Meiji Seika Pharma, Menlo, Mitsubishi, Neuroderm, Pfizer, Promius/Dr. Reddy’s Laboratories, Theravance, and Verrica, Stefan Schreiber Consultant of: AbbVie, Arena, BMS, Biogen, Celltrion, Celgene, IMAB, Gilead, MSD, Mylan, Pfizer, Fresenius, Janssen, Takeda, Theravance, provention Bio, Protagonist and Falk, Weibin Bao Shareholder of: Novartis, Employee of: Novartis, Kwaku Marfo Shareholder of: Novartis, Employee of: Novartis, Hanno Richards Shareholder of: Novartis, Employee of: Novartis, Luminita Pricop Shareholder of: Novartis, Employee of: Novartis, Abhijit Shete Shareholder of: Novartis, Employee of: Novartis, Jorge Safi Shareholder of: Novartis, Employee of: Novartis, Philip J Mease Grant/research support from: Abbott, Amgen, Biogen Idec, BMS, Celgene Corporation, Eli Lilly, Novartis, Pfizer, Sun Pharmaceutical, UCB – grant/research support, Consultant of: Abbott, Amgen, Biogen Idec, BMS, Celgene Corporation, Eli Lilly, Novartis, Pfizer, Sun Pharmaceutical, UCB – consultant, Speakers bureau: Abbott, Amgen, Biogen Idec, BMS, Eli Lilly, Genentech, Janssen, Pfizer, UCB – speakers bureau

Weakly-Supervised Convolutional Neural Network Architecture for Predicting Protein-DNA Binding

Although convolutional neural networks (CNN) have outperformed conventional methods in predicting the sequence specificities of protein-DNA binding in recent years, they do not take full advantage of the intrinsic weakly-supervised information of DNA sequences that a bound sequence may contain multiple TFBS(s). Here, we propose a weakly-supervised convolutional neural network architecture (WSCNN), combining multiple-instance learning (MIL) with CNN, to further boost the performance of predicting protein-DNA binding. WSCNN first divides each DNA sequence into multiple overlapping subsequences (instances) with a sliding window, and then separately models each instance using CNN, and finally fuses the predicted scores of all instances in the same bag using four fusion methods, including Max, Average, Linear Regression, and Top-Bottom Instances. The experimental results on in vivo and in vitro datasets illustrate the performance of the proposed approach. Moreover, models built on in vitro data using WSCNN can predict in vivo protein-DNA binding with good accuracy. In addition, we give a quantitative analysis of the importance of the reverse-complement mode in predicting in vivo protein-DNA binding, and explain why not directly use advanced pooling layers to combine MIL with CNN, through a series of experiments.

MSIT: Malonylation Sites Identification Tree

Aims:

Post-Translational Modifications (PTMs), which include more than 450 types, can be regarded as the fundamental cellular regulation.

Background:

Recently, experiments demonstrated that the lysine malonylation modification is a significant process in several organisms and cells. Meanwhile, malonylation plays an important role in the regulation of protein subcellular localization, stability, translocation to lipid rafts and many other protein functions.

Objective:

Identification of malonylation will contribute to understanding the molecular mechanism in the field of biology. Nevertheless, several existing experimental approaches, which can hardly meet the need of the high speed data generation, are expensive and time-consuming. Moreover, some machine learning methods can hardly meet the high-accuracy need in this issue.

Methods:

In this study, we proposed a method, named MSIT that means malonylation sites identification tree, utilized the amino acid residues and profile information to identify the lysine malonylation sites with the tree structural neural network in the peptides sequence level.

Methods:

The proposed algorithm can get 0.8699 of F1 score and 89.34% in true positive ratio in E. coli. MSIT outperformed existing malonylation site identification methods and features on different species datasets.

Conclusion:

Based on these measures, it can be demonstrated that MSIT will be helpful in identifying candidate malonylation sites.

1410Clinical predictors of NT-proBNP response to early initiation of sacubitril/valsartan after hospitalisation for decompensated heart failure: An analysis of the TRANSITION study

Background

NT-proBNP has diagnostic and prognostic value in patients with heart failure (HF). Compared with enalapril, sacubitril/valsartan (S/V) significantly reduced NT-proBNP within 1 week (wk) of administration and reduced HF re-hospitalisation in patients with acute decompensated HF (ADHF) in PIONEER-HF. Identification of predictors of NT-proBNP reduction with S/V could aid prognostication following hospitalisation.

Methods

TRANSITION (NCT02661217) is an open label study in stabilised ADHF patients with HFrEF that compared S/V initiation pre- versus post-discharge (within 2 wk of discharge). Baseline NT-proBNP was measured at randomisation in both S/V groups (n=950). Clinical predictors of favourable response of NT-proBNP to S/V therapy (defined as reduction to <1000 pg/ml or >30% reduction vs. baseline) were studied at discharge, 4 wk and 10 wk post-randomisation.

Results

Median NT-proBNP at randomisation was similar in patients with S/V started pre- and post-discharge (1919 vs 1659 pg/ml). In patients receiving S/V in-hospital, NT-proBNP was reduced by 28% at discharge, compared to a 3% reduction in patients receiving optimised standard of care (between group p<0.001). A favorable response was reached in 46% vs 18% patients at discharge, 46% vs 42% at 4 weeks and 51% vs 48% at 10 weeks in pre- vs post-discharge groups. (Figure 1). Predictors of favourable NT-proBNP response to S/V at discharge were hypertension and shorter time from admission to first S/V dose. At 4 wk after randomisation, NT-proBNP was reduced similarly in patients started on S/V pre- and post-discharge. When the two S/V initiation groups were combined, predictors of favorable NT-proBNP response at 4 wk were higher initial dose of S/V (≥49/51 mg b.i.d.), higher baseline levels of NT-proBNP, de novo HF hospitalisation, ACEI/ARB naïve, lower baseline creatinine, no atrial fibrillation (AFib), no prior myocardial infarction (MI). A further reduction in NT-proBNP was seen at 10 wk post-randomisation in patients started on S/V pre- and post-discharge (38% vs 34%, between group p=0.250). Predictors of favourable NT-proBNP response to S/V were similar at 4 wk and 10 wk post-randomisation.

Conclusion

In-hospital initiation of sacubitril/valsartan shortly after stabilisation was associated with a prompt improvement of NT-proBNP already at discharge, whereas higher baseline levels of NT-proBNP, higher starting dose, absence of AFib and MI history, de novo HF and ACEI/ARB naïve status were associated with favourable NT-proBNP response in the vulnerable phase after discharge.

Acknowledgement/Funding

The TRANSITION study was funded by Novartis

P773Initiation of sacubitril/valsartan and optimisation of evidence-based heart failure therapies after hospitalisation for acute decompensated heart failure: An analysis of the TRANSITION study

Background

Optimisation of chronic heart failure (HF) therapy remains the key strategy to improve outcomes after hospitalisation for acute decompensated HF (ADHF) with reduced ejection fraction (HFrEF). Initiation and uptitration of disease-modifying therapies is challenging in this vulnerable patient population. We aimed to describe the patterns of treatment optimisation including sacubitril/valsartan (S/V) in the TRANSITION study.

Methods

TRANSITION (NCT02661217) was a randomised, open-label study comparing S/V initiation pre- vs. post-discharge (1–14 days) in patients admitted for ADHF after haemodynamic stabilisation. The primary endpoint was the proportion of patients achieving 97/103 mg S/V twice daily (bid) at 10 weeks post-randomisation. Up-titration of S/V was as per label. Information on dose of S/V and on the use of concomitant HF medication was collected at each study visit up to week 26.

Results

A total of 493 patients received at least one dose of S/V in the pre-discharge arm and 489 patients in the post-discharge arm. One month after randomisation, 45% of patients in the pre-d/c arm vs. 44% in the post-discharge arm used 24/26 mg bid starting dose and 42% vs. 40% were on 49/51 mg S/V bid, respectively. At week 10, 47% of patients had achieved the target dose in the pre-discharge arm vs. 51% in the post-discharge arm. At the end of the follow-up at 26 weeks, the proportion of patients on S/V target dose further increased to 53% in the pre-discharge and 61% in the post-discharge arm (Figure 1). At week 10, the mean dose of S/V was 132 mg in the pre-discharge arm and 136 mg in the post-discharge arm, and at week 26, it was 140 mg and 147 mg, respectively.

Before hospital admission, 52% and 54% of the patients received a beta-blocker (BB) in the pre-discharge and post-discharge group, respectively, and 42% in both arms received a mineralcorticoid receptor antagonist (MRA). At time of discharge, 68% and 71%% of the patients received a BB and 68% and 65% an MRA, in the pre-discharge and post-discharge groups, respectively. These proportions remained stable to week 10 and week 26.

Uptitration of sacubitril/valsartan

Conclusions

In the vulnerable post-ADHF population, initiation of S/V and up-titration to target dose was feasible within 10 weeks in half of the patients alongside with a 20% increase in the use of other disease-modifying medications that remained stable through the end of the 6-month follow-up.

Acknowledgement/Funding

The TRANSITION study was funded by Novartis

P1637Rehospitalisations during 26 weeks of follow up from initiation of sacubitril/valsartan after acute decompensated heart failure: An analysis of the TRANSITION study

Background

Patients with acute decompensated heart failure (ADHF) are at high risk of recurrent hospitalisations and death. In-hospital initiation of sacubitril/valsartan (S/V) reduced the risk for HF re-hospitalisation by 44% compared to enalapril in the PIONEER-HF study during the 8-week follow-up period. We aimed to describe the pattern of readmissions and their causes in the TRANSITION study, which randomised participants to pre-discharge or post-discharge initation of S/V.

Methods

TRANSITION (NCT02661217) was a randomised, open-label study comparing S/V initiation pre- vs. post-discharge (1–14 days) in haemodynamically stabilised patients with HF with reduced ejection fraction, admitted for ADHF. The primary endpoint was the proportion of patients achieving 97/103 mg S/V twice daily at 10 weeks post-randomisation. Information on rehospitalisation was collected throughout the study up to 26 weeks.

Results

A total of 493 patients received S/V in the pre-discharge arm and 489 patients in the post-discharge arm. Readmissions due to any cause were reported in 9.7% and 18.1% in the pre-discharge arm vs. 10.6% and 21.3% in the post-discharge arm within 30 days, and 10 weeks respectively. During the 26-weeks follow-up, all-cause readmission was reported in 34.5% of patients in the pre-discharge arm vs. 34.6% in the post-discharge arm. Median time to first rehospitalisation was 67 days in the pre-discharge arm (IQR: 26–110 days) and 50 days (IQR: 23–108 days) in the post-discharge arm. At least one HF hospitalisation was reported in 7.5% of patients in the pre-discharge arm and 7.4% in the post-discharge arm during 10 weeks and in 11.8% and 12.3% of patients, respectively, during 26 weeks of follow-up. Median duration of HF readmission was 7 days (IQR: 4–11 days) in the pre-discharge group and 6.5 days (IQR: 6.5–10 days) in the post-discharge arm. In total 2.6% and 5.5% patients in pre-discharge arm and 3.9% and 7% in the post-discharge arm visited an emergency room during 10 weeks and 26 weeks, respectively.

Conclusions

Initiation of S/V in patients hospitalised for ADHF either before or shortly after discharge, results in comparable rates of all cause and HF rehospitalisations, as well as emergency room visits without hospital admission over the 26 week follow-up period. HF re-hospitalisations rates at 10 weeks in TRANSITION are in line with the 8% in S/V arm reported in PIONEER-HF during the 8-weeks follow-up.

Acknowledgement/Funding

The TRANSITION study was funded by Novartis

MiR-99a-5p regulates proliferation, migration and invasion abilities of human oral carcinoma cells by targeting NOX4

Previous research has showed that miR-99a-5p was a tumor suppressor. The aim of our study was to explore the effect of miR-99a-5p on the vitality and proliferation, migration together with the invasion of oral tumor cells via inhibiting the expression of NOX4. QRT-PCR and Western blot were applied to examine the expression level of miR-99a-5p and NOX4 in human oral tumorous and adjacent tissues. Dual luciferase reporter gene assay was applied to confirm that miR-99a-5p negatively regulated directly on NOX4 in TSCC1 cells. Cell transfection and lentiviral vectors were used to up-regulate expression of miR-99a-5p and NOX4, respectively. Cell proliferation, cell cycle, apoptosis and invasion along with the migration in different groups were assessed using MTT assay, colony formation assay, the flow cytometry, transwell assay and the wound healing assay, respectively. MiR-99a-5p was under-expressed in human oral tumor, while NOX4 was over-expressed. There was a negative relationship between miR-99a-5p and NOX4. Up-regulating miR-99a-5p or down-regulating NOX4 suppressed the vitality, proliferation, migration together with invasion of TSCC1 cells. MiR-99a-5p affected the vitality and proliferation, migration together with the invasion of oral tumor cells through targeting NOX4.

Abstract P6-18-05: First-line ribociclib + endocrine therapy in hormone receptor-positive, HER2-negative advanced breast cancer: A pooled efficacy analysis

Background: In three separate Phase III randomized, placebo-controlled trials, ribociclib (RIB; cyclin-dependent kinase 4/6 inhibitor) + various endocrine therapy (ET) partners prolonged progression-free survival (PFS) vs placebo (PBO) + ET in patients (pts) with hormone receptor-positive (HR+), HER2-negative (HER2–) advanced breast cancer (ABC). Here we further evaluate the efficacy of RIB-based regimens of interest (i.e. with a non-steroidal aromatase inhibitor [NSAI] or fulvestrant [FUL]) in pts who were ET-naïve in the ABC setting, using pooled data from three Phase III trials: MONALEESA (ML)-2 (NCT01958021; all pts), ML-3 (NCT02422615; no prior ET for ABC subgroup only), and ML-7 (NCT02278120; RIB + NSAI subgroup only).

Methods: Postmenopausal pts with no prior ET for ABC received RIB (600 mg/day; 3-weeks-on/1-week-off) or PBO + either letrozole (2.5 mg/day) in ML-2 or FUL (500 mg every 28 days, with an additional dose on Day 15 of Cycle 1) in ML-3. In ML-7, premenopausal pts with no prior ET and ≤1 line of chemotherapy for ABC received RIB or PBO + goserelin (3.6 mg every 28 days) + NSAI (anastrozole [1 mg/day]/letrozole [2.5 mg/day]). The primary endpoint of all three trials was locally assessed PFS. Secondary endpoints included overall response rate (ORR), clinical benefit rate (CBR), and duration of response (DoR; ML-3 and -7). DoR was an exploratory endpoint in ML-2.

Results: Data were pooled for 820 pts treated with RIB + ET (ML-2: n=334; ML-3: n=238; ML-7: n=248) and 710 pts treated with PBO + ET (ML-2: n=334; ML-3: n=129; ML-7: n=247). As of the data cutoffs (ML-2: January 2, 2017; ML-3: November 3, 2017; ML-7: August 20, 2017), in the RIB + ET vs PBO + ET arms, 385 (47%) vs 234 (33%) pts remained on-treatment; the most common reason for discontinuation was disease progression (n=292 [36%] vs n=391 [55%]). In this pooled analysis, median PFS was prolonged for RIB + ET vs PBO + ET, with a hazard ratio of 0.570 (95% confidence interval [CI] 0.491–0.662); median PFS was 25.3 months (95% CI 23.9–29.6) vs 15.6 months (95% CI 14.4–16.9), respectively. Consistent PFS benefit for RIB + ET vs PBO + ET was observed across pt subgroups, including ECOG performance status, age, race, or presence/absence of liver and/or lung metastases or bone-only disease. Among all pts in the pooled analysis, the ORR was 41% for RIB + ET vs 28% for PBO + ET and the CBR was 79% vs 70%, respectively. In pts with measurable disease at baseline (RIB + ET: n=639; PBO + ET: n=542), the ORR was 51% for RIB + ET vs 37% for PBO + ET and the CBR was 79% vs 68%, respectively. In the RIB + ET vs PBO + ET arms, the median DoR was 26.7 months vs 20.0 months. A decrease in best percentage change from baseline in the sum of longest diameters per RECIST was observed in 86% of pts receiving RIB + ET vs 73% of pts receiving PBO + ET.

Conclusions: RIB in combination with various ET partners demonstrates improved clinical outcomes vs PBO + ET across a broad population of pts with HR+, HER2– ABC. These data provide further support for the use of RIB-based combinations in pre- and postmenopausal pts with HR+, HER2– ABC who have received no prior ET for advanced disease.

Citation Format: Tripathy D, Hortobagyi G, Chan A, Im S-A, Chia S, Yardley D, Esteva FJ, Hurvitz S, Kong O, Bao W, Rodriguez Lorenc K, Diaz-Padilla I, Slamon DJ. First-line ribociclib + endocrine therapy in hormone receptor-positive, HER2-negative advanced breast cancer: A pooled efficacy analysis [abstract]. In: Proceedings of the 2018 San Antonio Breast Cancer Symposium; 2018 Dec 4-8; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2019;79(4 Suppl):Abstract nr P6-18-05.

Abstract P6-18-15: Ribociclib + endocrine therapy in hormone receptor-positive, HER2-negative advanced breast cancer: A pooled safety analysis

Background: In Phase III trials, ribociclib (RIB; cyclin-dependent kinase 4/6 inhibitor) + various endocrine therapy (ET) partners has demonstrated significantly prolonged progression-free survival vs placebo (PBO) + ET in patients (pts) with hormone receptor-positive (HR+), HER2-negative (HER2–) advanced breast cancer (ABC). Here we further evaluate the safety of RIB-based regimens of interest for the proposed indication (i.e. with a non-steroidal aromatase inhibitor [NSAI] or fulvestrant [FUL]) using pooled data from three Phase III trials (MONALEESA [ML]-2 [NCT01958021], -3 [NCT02422615], and -7 [NCT02278120]).

Methods: Postmenopausal pts with HR+, HER2– ABC received RIB (600 mg/day; 3-weeks-on/1-week-off) or PBO + letrozole (LET; 2.5 mg/day; ML-2 [no prior ET for ABC]) or FUL (500 mg, Days 1 and 15 of Cycle 1, then Day 1 of every cycle thereafter; ML-3; no or ≤1 prior line of ET for ABC]). Premenopausal pts (ML-7; no prior ET and ≤1 chemotherapy for ABC]) received RIB or PBO + anastrozole (1 mg/day)/LET (2.5 mg/day) + goserelin (3.6 mg every 28 days). Adverse events (AEs) were characterized per Common Terminology Criteria for Adverse Events v4.03; safety analyses included time to first event, duration of event, and rate of associated RIB/PBO discontinuations.

Results: Data for 1883 pts were pooled; 1065 pts received RIB + ET and 818 pts received PBO + ET (median exposure to study treatment: 17 and 13 months, respectively). Exposure-adjusted incidence rates for AEs of special interest were 561 and 131 per 100 pt-years in the RIB and PBO arms, respectively. The most common all-causality Grade 3/4 AEs (≥10% in any arm; RIB vs PBO) were neutropenia (59% vs 2%), leukopenia (18% vs 1%), and hypertension (13% vs 13%). A new Fridericia's corrected QT interval (QTcF) &gt;480 ms occurred in (n/N) 52/1054 (5%) vs 11/814 (1%) pts in the RIB vs PBO arms; a new QTcF &gt;500 ms occurred in 14/1054 (1%) vs 1/814 (&lt;1%) pts. Median time to first event for Grade ≥2 neutropenia, elevated alanine aminotransferase (ALT) and/or aspartate aminotransferase (AST), and QTc prolongation in the RIB arm was 2, 12, and 2 weeks, respectively; median duration of first Grade ≥2 event was 4, 4, and 2 weeks. In the RIB arm vs PBO arms, 7% vs 3% of pts discontinued study treatment due to AEs; common all-grade AEs leading to RIB/PBO discontinuation (≥2% in any arm) were elevated ALT (4% vs &lt;1%) and elevated AST (2% vs 1%). Discontinuation due to QT prolongation occurred in 4 pts in the RIB arm and 2 in the PBO arm (both &lt;1%). All-grade serious AEs occurred in 25% of pts in the RIB arm vs 15% of pts in the PBO arm.

Conclusions: RIB in combination with various ET partners continues to demonstrate a predictable and manageable tolerability profile across a broad population of pts with HR+, HER2– ABC.

Citation Format: Burris HA, Chan A, Im S-A, Chia S, Tripathy D, Esteva FJ, Campone M, Bardia A, Kong O, Bao W, Diaz-Padilla I, Rodriguez Lorenc K, Yardley DA. Ribociclib + endocrine therapy in hormone receptor-positive, HER2-negative advanced breast cancer: A pooled safety analysis [abstract]. In: Proceedings of the 2018 San Antonio Breast Cancer Symposium; 2018 Dec 4-8; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2019;79(4 Suppl):Abstract nr P6-18-15.

LAIPT: Lysine Acetylation Site Identification with Polynomial Tree

Post-translational modification plays a key role in the field of biology. Experimental identification methods are time-consuming and expensive. Therefore, computational methods to deal with such issues overcome these shortcomings and limitations. In this article, we propose a lysine acetylation site identification with polynomial tree method (LAIPT), making use of the polynomial style to demonstrate amino-acid residue relationships in peptide segments. This polynomial style was enriched by the physical and chemical properties of amino-acid residues. Then, these reconstructed features were input into the employed classification model, named the flexible neural tree. Finally, some effect evaluation measurements were employed to test the model’s performance.

HSCVFNT: Inference of Time-Delayed Gene Regulatory Network Based on Complex-Valued Flexible Neural Tree Model

Gene regulatory network (GRN) inference can understand the growth and development of animals and plants, and reveal the mystery of biology. Many computational approaches have been proposed to infer GRN. However, these inference approaches have hardly met the need of modeling, and the reducing redundancy methods based on individual information theory method have bad universality and stability. To overcome the limitations and shortcomings, this thesis proposes a novel algorithm, named HSCVFNT, to infer gene regulatory network with time-delayed regulations by utilizing a hybrid scoring method and complex-valued flexible neural network (CVFNT). The regulations of each target gene can be obtained by iteratively performing HSCVFNT. For each target gene, the HSCVFNT algorithm utilizes a novel scoring method based on time-delayed mutual information (TDMI), time-delayed maximum information coefficient (TDMIC) and time-delayed correlation coefficient (TDCC), to reduce the redundancy of regulatory relationships and obtain the candidate regulatory factor set. Then, the TDCC method is utilized to create time-delayed gene expression time-series matrix. Finally, a complex-valued flexible neural tree model is proposed to infer the time-delayed regulations of each target gene with the time-delayed time-series matrix. Three real time-series expression datasets from (Save Our Soul) SOS DNA repair system in E. coli and Saccharomyces cerevisiae are utilized to evaluate the performance of the HSCVFNT algorithm. As a result, HSCVFNT obtains outstanding F-scores of 0.923, 0.8 and 0.625 for SOS network and (In vivo Reverse-Engineering and Modeling Assessment) IRMA network inference, respectively, which are 5.5%, 14.3% and 72.2% higher than the best performance of other state-of-the-art GRN inference methods and time-delayed methods.

Mutli-Features Prediction of Protein Translational Modification Sites

Post translational modification plays a significiant role in the biological processing. The potential post translational modification is composed of the center sites and the adjacent amino acid residues which are fundamental protein sequence residues. It can be helpful to perform their biological functions and contribute to understanding the molecular mechanisms that are the foundations of protein design and drug design. The existing algorithms of predicting modified sites often have some shortcomings, such as lower stability and accuracy. In this paper, a combination of physical, chemical, statistical, and biological properties of a protein have been ulitized as the features, and a novel framework is proposed to predict a protein's post translational modification sites. The multi-layer neural network and support vector machine are invoked to predict the potential modified sites with the selected features that include the compositions of amino acid residues, the E-H description of protein segments, and several properties from the AAIndex database. Being aware of the possible redundant information, the feature selection is proposed in the propocessing step in this research. The experimental results show that the proposed method has the ability to improve the accuracy in this classification issue.

Randomised clinical trial: gastrointestinal events in arthritis patients treated with celecoxib, ibuprofen or naproxen in the PRECISION trial

AIM

To evaluate GI safety of celecoxib compared with 2 nonselective (ns) NSAIDs, as a secondary objective of a large trial examining multiorgan safety.

METHODS

This randomised, double-blind controlled trial analysed 24 081 patients. Osteoarthritis or rheumatoid arthritis patients, needing ongoing NSAID treatment, were randomised to receive celecoxib 100-200 mg b.d., ibuprofen 600-800 mg t.d.s. or naproxen 375-500 mg b.d. plus esomeprazole, and low-dose aspirin or corticosteroids if already prescribed. Clinically significant GI events (CSGIE-bleeding, obstruction, perforation events from stomach downwards or symptomatic ulcers) and iron deficiency anaemia (IDA) were adjudicated blindly.

RESULTS

Mean treatment and follow-up durations were 20.3 and 34.1 months. While on treatment or 30 days after, CSGIE occurred in 0.34%, 0.74% and 0.66% taking celecoxib, ibuprofen and naproxen. Hazard ratios (HR) were 0.43 (95% CI 0.27-0.68, P = 0.0003) celecoxib vs ibuprofen and 0.51 (0.32-0.81, P = 0.004) vs naproxen. There was also less IDA on celecoxib: HR 0.43 (0.27-0.68, P = 0.0003) vs ibuprofen; 0.40 (0.25-0.62, P < 0.0001) vs naproxen. Even taken with low-dose aspirin, fewer CSGIE occurred on celecoxib than ibuprofen (HR 0.52 [0.29-0.94], P = 0.03), and less IDA vs naproxen (0.42 [0.23-0.77, P = 0.005]). Corticosteroid use increased total GI events and CSGIE. H. pylori serological status had no influence.

CONCLUSIONS

Arthritis patients taking NSAIDs plus esomeprazole have infrequent clinically significant gastrointestinal events. Co-prescribed with esomeprazole, celecoxib has better overall GI safety than ibuprofen or naproxen at these doses, despite treatment with low-dose aspirin or corticosteroids.

Phylogeny of dermatophytes with genomic character evaluation of clinically distinct Trichophyton rubrum and T. violaceum

Trichophyton rubrum and T. violaceum are prevalent agents of human dermatophyte infections, the former being found on glabrous skin and nail, while the latter is confined to the scalp. The two species are phenotypically different but are highly similar phylogenetically. The taxonomy of dermatophytes is currently being reconsidered on the basis of molecular phylogeny. Molecular species definitions do not always coincide with existing concepts which are guided by ecological and clinical principles. In this article, we aim to bring phylogenetic and ecological data together in an attempt to develop new species concepts for anthropophilic dermatophytes. Focus is on the T. rubrum complex with analysis of rDNA ITS supplemented with LSU, TUB2, TEF3 and ribosomal protein L10 gene sequences. In order to explore genomic differences between T. rubrum and T. violaceum, one representative for both species was whole genome sequenced. Draft sequences were compared with currently available dermatophyte genomes. Potential virulence factors of adhesins and secreted proteases were predicted and compared phylogenetically. General phylogeny showed clear gaps between geophilic species of Arthroderma, but multilocus distances between species were often very small in the derived anthropophilic and zoophilic genus Trichophyton. Significant genome conservation between T. rubrum and T. violaceum was observed, with a high similarity at the nucleic acid level of 99.38 % identity. Trichophyton violaceum contains more paralogs than T. rubrum. About 30 adhesion genes were predicted among dermatophytes. Seventeen adhesins were common between T. rubrum and T. violaceum, while four were specific for the former and eight for the latter. Phylogenetic analysis of secreted proteases reveals considerable expansion and conservation among the analyzed species. Multilocus phylogeny and genome comparison of T. rubrum and T. violaceum underlined their close affinity. The possibility that they represent a single species exhibiting different phenotypes due to different localizations on the human body is discussed.

Novel human microbe-disease association prediction using network consistency projection

BACKGROUND

Accumulating biological and clinical reports have indicated that imbalance of microbial community is closely associated with occurrence and development of various complex human diseases. Identifying potential microbe-disease associations, which could provide better understanding of disease pathology and further boost disease diagnostic and prognostic, has attracted more and more attention. However, hardly any computational models have been developed for large scale microbe-disease association prediction.

RESULTS

In this article, based on the assumption that microbes with similar functions tend to share similar association or non-association patterns with similar diseases and vice versa, we proposed the model of Network Consistency Projection for Human Microbe-Disease Association prediction (NCPHMDA) by integrating known microbe-disease associations and Gaussian interaction profile kernel similarity for microbes and diseases. NCPHMDA yielded outstanding AUCs of 0.9039, 0.7953 and average AUC of 0.8918 in global leave-one-out cross validation, local leave-one-out cross validation and 5-fold cross validation, respectively. Furthermore, colon cancer, asthma and type 2 diabetes were taken as independent case studies, where 9, 9 and 8 out of the top 10 predicted microbes were successfully confirmed by recent published clinical literature.

CONCLUSION

NCPHMDA is a non-parametric universal network-based method which can simultaneously predict associated microbes for investigated diseases but does not require negative samples. It is anticipated that NCPHMDA would become an effective biological resource for clinical experimental guidance.

CIPPN: computational identification of protein pupylation sites by using neural network

Recently, experiments revealed the pupylation to be a signal for the selective regulation of proteins in several serious human diseases. As one of the most significant post translational modification in the field of biology and disease, pupylation has the ability to playing the key role in the regulation various diseases’ biological processes. Meanwhile, effectively identification such type modification will be helpful for proteins to perform their biological functions and contribute to understanding the molecular mechanism, which is the foundation of drug design. The existing algorithms of identification such types of modified sites often have some defects, such as low accuracy and time-consuming. In this research, the pupylation sites’ identification model, CIPPN, demonstrates better performance than other existing approaches in this field. The proposed predictor achieves Acc value of 89.12 and Mcc value of 0.7949 in 10-fold cross-validation tests in the Pupdb Database (http://cwtung.kmu.edu.tw/pupdb). Significantly, such algorithm not only investigates the sequential, structural and evolutionary hallmarks around pupylation sites but also compares the differences of pupylation from the environmental, conservative and functional characterization of substrates. Therefore, the proposed feature description approach and algorithm results prove to be useful for further experimental investigation of such modification’s identification.

Classification of Protein Structure Classes on Flexible Neutral Tree

Accurate classification on protein structural is playing an important role in Bioinformatics. An increase in evidence demonstrates that a variety of classification methods have been employed in such a field. In this research, the features of amino acids composition, secondary structure's feature, and correlation coefficient of amino acid dimers and amino acid triplets have been used. Flexible neutral tree (FNT), a particular tree structure neutral network, has been employed as the classification model in the protein structures' classification framework. Considering different feature groups owing diverse roles in the model, impact factors of different groups have been put forward in this research. In order to evaluate different impact factors, Impact Factors Scaling (IFS) algorithm, which aim at reducing redundant information of the selected features in some degree, have been put forward. To examine the performance of such framework, the 640, 1189, and ASTRAL datasets are employed as the low-homology protein structure benchmark datasets. Experimental results demonstrate that the performance of the proposed method is better than the other methods in the low-homology protein tertiary structures.