Predictors of mortality in patients with VA-extracorporeal membrane oxygenation

Background

Venoarterial extracorporeal membrane oxygenation (VA-ECMO) support is a lifesaving tool used in the treatment of cardiogenic shock, acute heart failure, or extracorporeal cardiopulmonary resuscitation (CPR). We report on a single center experience with ECMO and aim to identify the prognostic markers for in-hospital mortality and death at 72 hours after ECMO.

Methods

Between 2011 and 2019 we evaluated 131 patients, who received ECMO. Collected data was analyzed to identify baseline characteristic, outcomes including clinical variables predictive of poor outcome.

Results

The mean age was 62.5 years, 67.2% were male patients, with prior CPR in 61.8%. The annual number of VA-ECMO procedures steadily increased, whereas in-hospital mortality is decreasing. Within the total cohort, the indication for VA-ECMO was cardiac arrest in 19.1%, acute coronary syndrome in 41.2%, acute heart failure in 23.7%, and myocarditis in 10.7%.

Overall in-hospital mortality was 58.8%. Multivariate logistic regression model revealed presence of malignancy, history of revascularization, duration of cardiac arrest, and low BMI as independent predictors for mortality in 72 hours after ECMO (table). On the other hand, predictors of in-hospital mortality were prior congestive heart failure, male, and history of malignancy. The C-statistic for discriminating mortality in 72 hours after ECMO with the duration of cardiac arrest was 0.67 (figure).

Conclusions

Although the use of ECMO as a last line in the treatment of critical patients measures constitutes an important improvement in their care; with 41.2% overall survival; patient selection and timing of ECMO initiation remains challenging. The importance of consideration for ECMO use earlier in course of illness rather than later.

Funding Acknowledgement

Type of funding source: None

Intense cardiac-targeted small extracellular vesicles-mediated delivery of RAGE siRNA attenuates inflammation in rat myocarditis model

Background

Despite advances in the field and new therapeutics being developed, heart disease remains the leading cause of mortality worldwide. Small extracellular vesicles (sEV) are 30–150 nm in diameter and ferry RNA and proteins to cells. sEV are a natural carrier of many signaling molecules and play an important role in heart disease. However, the high probability of off-target effects associated with these carriers is a major barrier to translation into clinical application.

Purpose

Our aim was to evaluate whether sEVs engineered to express potent cardiac targeting peptides (CTPs) could deliver siRNA to the heart and exert a therapeutic effect.

Methods

We use vectors encoding LAMP2B (sEV) or CTP-LAMP2B (PsEV) into HEK 293 cells expressing. sEVs were purified from culture media of HEK 293 cells by serial centrifugation followed by tangential flow filtration (TFF) system. sEV and PsEV were loaded with siRNAs by Exo-Fect™ exosome transfection reagent, and were treated into H9C2 rat cardiomyocyte. TNF-alpha were then added to the cells to induce inflammation. And sEVs were intravenously injected into myocarditis rat. Inflammation factors of in vitro and in vivo inflammation model were identified by western blot. Echocardiographic examination was also performed in rat.

Results

The successful development of PsEVs was analyzed by Western blot and TEM. We observed a 4 fold increase from that of the previously developed sEVs in H9C2 cells and a 200% increase in cardiac-specific delivery without toxicity in rat model. AGE is involved in proinflammatory/pro-apoptotic processes. To block RAGE, we loaded RAGE siRNA (siRAGE) in sEV that had high expression of PsEV. Characteristics of sEVs were maintained despite siRNA load.In inflammatory cell models and rat disease models, PsEV-treated groups significantly reduced molecular levels associated with inflammatory responses such as RAGE, IL-6, TNF-alpha, COX2, HMGB1, and p-p65 / p65.

Conclusions

Our results suggest that PsEV can potentially serve as a treatment delivery vehicle for heart disease.

Scheme

Funding Acknowledgement

Type of funding source: None

Prediction of clinically actionable genetic alterations from colorectal cancer histopathology images using deep learning

BACKGROUND

Identifying genetic mutations in cancer patients have been increasingly important because distinctive mutational patterns can be very informative to determine the optimal therapeutic strategy. Recent studies have shown that deep learning-based molecular cancer subtyping can be performed directly from the standard hematoxylin and eosin (H&E) sections in diverse tumors including colorectal cancers (CRCs). Since H&E-stained tissue slides are ubiquitously available, mutation prediction with the pathology images from cancers can be a time- and cost-effective complementary method for personalized treatment.

AIM

To predict the frequently occurring actionable mutations from the H&E-stained CRC whole-slide images (WSIs) with deep learning-based classifiers.

METHODS

A total of 629 CRC patients from The Cancer Genome Atlas (TCGA-COAD and TCGA-READ) and 142 CRC patients from Seoul St. Mary Hospital (SMH) were included. Based on the mutation frequency in TCGA and SMH datasets, we chose APC, KRAS, PIK3CA, SMAD4, and TP53 genes for the study. The classifiers were trained with 360 × 360 pixel patches of tissue images. The receiver operating characteristic (ROC) curves and area under the curves (AUCs) for all the classifiers were presented.

RESULTS

The AUCs for ROC curves ranged from 0.693 to 0.809 for the TCGA frozen WSIs and from 0.645 to 0.783 for the TCGA formalin-fixed paraffin-embedded WSIs. The prediction performance can be enhanced with the expansion of datasets. When the classifiers were trained with both TCGA and SMH data, the prediction performance was improved.

CONCLUSION

APC, KRAS, PIK3CA, SMAD4, and TP53 mutations can be predicted from H&E pathology images using deep learning-based classifiers, demonstrating the potential for deep learning-based mutation prediction in the CRC tissue slides.

Melatonin regulates mitochondrial function and biogenesis during rat dental papilla cell differentiation

OBJECTIVE

The aim of this study was to investigate the effect of melatonin on mitochondria of dental papilla cells (DPCs) during the odontogenic differentiation process.

MATERIALS AND METHODS

Primary DPCs were obtained from the first molar dental papilla of neonatal rats and cultured in osteogenic (OS) or basal medium supplemented with melatonin at different concentrations (0, 1 pM, 0.1 nM, 10 nM, and 1 μM) for differentiation in vitro. Effects of melatonin on differentiation, mitochondrial respiratory function, and mitochondrial biogenesis of DPCs were analyzed.

RESULTS

Upon odontogenic induction, Alkaline phosphatase (ALP) activity, dentin sialophosphoprotein (DSPP), and dentin matrix protein (DMP1) expression were significantly enhanced, with a peaked expression at 10 nM of melatonin treatment. During DPCs differentiation, 10 nM melatonin could significantly induce the increase of intracellular Adenosine triphosphate (ATP), the decrease of the oxidized form of nicotinamide adenine dinucleotide (NAD+)/NADH ratio and reactive oxygen species (ROS). The mRNA and protein levels of peroxisome proliferator-activated receptor gamma coactivator 1-alpha (PGC-1α), nuclear respiratory factor 1 (NRF-1), and mitochondrial transcription factor A (TFAM) were significantly increased, and the peak level of expression was found in cells treated with 10 nM of melatonin. Furthermore, the mitochondria DNA (mtDNA) copy number was significantly decreased during DPCs differentiation.

CONCLUSIONS

These findings suggest that melatonin can promote the differentiation of rat DPCs and regulate mitochondrial energy metabolism, ROS scavenging, and mitochondrial biogenesis.

Association between immunotherapy biomarkers and glucose metabolism from F-18 FDG PET

OBJECTIVE

To assess associations between parameters derived from F-18 fluorodeoxyglucose (FDG) positron emission tomography (PET) and mRNA expression levels of immune checkpoint biomarkers such as programmed death receptor 1 (PD-1), programmed death-ligand 1 (PD-L1), cytotoxic T-lymphocyte antigen 4 (CTLA-4) as well as tumor mutation burden (TMB) in non-small cell lung cancer (NSCLC) patients.

PATIENTS AND METHODS

Integrated data were downloaded from Genomic Data Common Data Portal. Clinical, mRNA-seq, and whole exome-seq data of lung adenocarcinoma and squamous cell carcinoma from The Cancer Genome Atlas (TCGA) database were analyzed. TMB was defined as the total number of somatic missense mutations per megabase of the genome examined. Expression levels of PD-1, PD-L1, CTLA4 mRNA and TMB were collected. Correlations between imaging parameters of glucose metabolism and the expression levels of genomic biomarkers from cancers were evaluated. Bonferroni correction (adjusted p<0.0027) was applied to reduce type 1 error.

RESULTS

Of 31 NSCLC cases, 11 cases were adenocarcinoma (LUAD) and 20 were squamous cell carcinoma (LUSC). In linear regression analysis, texture parameters such as low gray-level run emphasis (LGRE, R2=0.48, p<0.0001), short run low gray-level emphasis (SRLGE, R2=0.45, p<0.0001) and long run low gray-level emphasis (LRLGE, R2=0.41, p=0.0001) derived from gray-level run length matrix (GLRLM) showed remarkable correlation with PD-L1 mRNA expression. Expression of PD-1, CTLA-4, and TMB failed to show any significant correlation with parameters of the F-18 FDG PET/CT.

CONCLUSIONS

Texture parameters derived from PET, known to indicate glucose uptake distribution, were correlated with expression of PD-L1 mRNA but not with expression of PD-1, CTLA-4 and TMB. Thus, tumoral heterogeneity could be a surrogate marker for the identification of PD-L1 level in NSCLC.

The Lq- NORM LEARNING FOR ULTRAHIGH-DIMENSIONAL SURVIVAL DATA: AN INTEGRATIVE FRAMEWORK

In the era of precision medicine, survival outcome data with high-throughput predictors are routinely collected. Models with an exceedingly large number of covariates are either infeasible to fit or likely to incur low predictability because of overfitting. Variable screening is key in identifying and removing irrelevant attributes. Recent years have seen a surge in screening methods, but most of them rely on some particular modeling assumptions. Motivated by a study on detecting gene signatures for multiple myeloma patients' survival, we propose a model-free L q -norm learning procedure, which includes the well-known Cramér-von Mises and Kolmogorov criteria as two special cases. The work provides an integrative framework for detecting predictors with various levels of impact, such as short- or long-term impact, on censored outcome data. The framework naturally leads to a scheme which combines results from different q to reduce false negatives, an aspect often overlooked by the current literature. We show that our method possesses sure screening properties. The utility of the proposal is confirmed with simulation studies and an analysis of the multiple myeloma study.

SAT0045 DISTRIBUTION AND SEVERITY OF PERIODONTITIS PREDICTS PROGRESSION TO INFLAMMATORY ARTHRITIS IN ANTI-CCP POSITIVE AT-RISK INDIVIDUALS WITHOUT CLINICAL SYNOVITIS

Background:

The prevalence of periodontal disease and the citrullinating bacteriumPorphyromonas gingivalisare increased in anti-CCP positive individuals at-risk of rheumatoid arthritis (RA) (1). This suggests periodontal inflammation may have an important role in the initiation and development of RA. Despite significant interest in the role of the periodontium and other mucosal sites in the initiation of RA-related autoimmunity, the influence of mucosal inflammation on progression to inflammatory arthritis (IA) in at-risk individuals remains unclear.

Objectives:

To investigate the association between periodontitis and progression to inflammatory arthritis in anti-CCP+ at-risk individuals without synovitis.

Methods:

Anti-CCP positive individuals with musculoskeletal symptoms but no clinical synovitis (CCP+ at-risk) were recruited as part of a national prospective cohort study. Comprehensive periodontal examination was performed at baseline by a dentist; six sites per tooth were assessed for clinical attachment level (CAL), pocket depth (PD) and bleeding on probing (BOP). Periodontal disease sites (PDD) were defined as CAL ≥2mm and PD ≥4mm. The distribution of PDD was classified in line with recent guidelines (2). The severity i.e. total burden of periodontal inflammation, was quantified at patient level using the periodontal inflamed surface area (PISA) index(3). CCP+ at-risk were monitored for progression to IA. Multivariable Cox regression was used to assess the effect of PDD distribution and PISA on progression to IA.

Results:

126 CCP+ at-risk underwent full periodontal examination and were followed up for median 23.4 months (range 0.6 – 56.8 months). Mean age was 49 years, 86 (68%) were females. At baseline, 42(33%) subjects had no PDD, 51(40%) had localised PDD (<30% teeth with one or more PDD site) and 33 (26%) had generalised PDD (≥ 30% of teeth with one or more PDD site). Mean (SD) PISA for all subjects was 267(319)mm2. 31 subjects (25%) progressed to IA after median of 12.6 months (range 0.6 – 49.5 months). Progression to IA was significantly higher in subjects with localised PDD compared with those without PDD (33% vs 16%, HR (95% CI) 2.45 (1.02, 5.94) p=0.02), figure 1. Interestingly, this association was not seen in subjects with generalised PDD (19% progression, HR 0.68 (0.20, 2.32). In addition, severity (i.e. total burden) of periodontal inflammation (PISA) was not significantly predictive of progression to IA alone (HR 1.001 (0.999-1.002), p=0.08). However, when adjusting for distribution of PDD, PISA was significantly associated with progression to IA (HR 1.0016 (1.0003- 1.003), p=0.00163).

Conclusion:

Periodontal inflammation predicts progression to IA in CCP+ at-risk individuals without clinical synovitis. The severity (i.e. total burden) of periodontitis appears to be particularly predictive of progression to IA in patients with localised periodontitis. These data suggest periodontitis may be an important factor in the development of RA and provide rationale for periodontal intervention with the aim of arthritis prevention in at-risk individuals.

References:

[1] Mankia K et al, JAMA Network Open(2019)

[2] Caton J et al, J Clin Periodontol(2018)

[3] Nesse W et al, J Clin Periodontol(2008)

Disclosure of Interests :

Kulveer Mankia: None declared, Zhain Mustufvi: None declared, Jing Kang: None declared, Aradhna Tugnait: None declared, Robert Letton: None declared, Laurence Duquenne: None declared, Alastair Speirs: None declared, Val Clerehugh: None declared, Deirdre Devine: None declared, Paul Emery Grant/research support from: AbbVie, Bristol-Myers Squibb, Merck Sharp & Dohme, Pfizer, Roche (all paid to employer), Consultant of: AbbVie (consultant, clinical trials, advisor), Bristol-Myers Squibb (consultant, clinical trials, advisor), Lilly (clinical trials, advisor), Merck Sharp & Dohme (consultant, clinical trials, advisor), Novartis (consultant, clinical trials, advisor), Pfizer (consultant, clinical trials, advisor), Roche (consultant, clinical trials, advisor), Samsung (clinical trials, advisor), Sandoz (clinical trials, advisor), UCB (consultant, clinical trials, advisor)

SAT0575 POSITIVE QuantiFERON-TB GOLD TEST AND SEROCONVERSION OF QunatiFERON-TB GOLD TEST IS ASSOCIATED WITH INCREASED RISK OF THE DEVELOPMENT OF ACTIVE TUBERCULOSIS IN PATIENTS WITH ANKYLOSING SPONDYLITIS: RESULTS FROM A REAL-WORLD DATA OVER 20 YEARS

Background:

With the promising efficacy and the prevalent use of anti-tumor necrosis factor-α (TNF) agents in managing AS, the risk for reactivation of latent tuberculosis infection (LTBI) still is a concern. Although guidelines include the screening and treatment of LTBI prior to the initiation of anti-TNF agent by QuantiFERON-TB Gold (QFT-G) or tuberculin test, there is a lack of evidence whether treatment of LTBI before initiation of anti-TNF agent may reduce the risk of reactivation to the same as LTBI patient without anti-TNF agent or anti-TNF agent users without LTBI. Furthermore, evidence on the need for follow-up testing and the association between seroconversion and the development of active tuberculosis is also limited.

Objectives:

This study aims to investigate the real-world impact of QFT-G test on the development of active tuberculosis in patients with AS.

Methods:

This retrospective study investigated 2,930 patients who had a diagnosis of AS and conducted QFT-G testing during the period of March 1998 to June 2019. 191 patients with history of treatment for LTBI or acute tuberculosis prior to the first QFT-G test and 157 patients whose hospital visits or prescription was less than 3 were excluded. Observational period was defined from the firs QFT-G test to the last hospital visit of development of active tuberculosis. The screening for development of active tuberculosis was conducted by reviewing the diagnosis, prescription of anti-tuberculosis medication, chest images and electronic medical record. Treatment of LTBI was defined when a patient was prescribed isoniazid for at least 220 for 12 months, rifampin for at least 90 days for 6 months, or concurrently prescribed isoniazid and rifampin for at least 70 days for 4 months. Wilcoxon rank-sum test, chi-square test and cox-proportional hazard analysis were performed.

Results:

A total of 2687 patients (median age 32.7 years, 78.4% male, anti-TNF agent user 16.7%) were included. Baseline QFT-G was positive in 426 (20.3%) patients, and 15 active tuberculosis was observed [Incidence rate 1.5/1000 person years (PY)]. Compared with baseline QFT-G (-) patients, baseline QFT-G (+) patients were older (41.2 years vs. 31.3 years, p<0.001) and they were accompanied with more active tuberculosis (4.4/1000PY vs. 1.0/1000PY) despite the less usage of anti-TNF agents (38.5% vs 45.8%, p=0.006). The observational period, sex, and medication utilization pattern except anti-TNF agent were similar between two groups. Multivariable analysis showed that QFT-G (+) test increases the risk of active tuberculosis more than 10 times [adjusted hazard ration 17.0, 95% confidence interval (CI) 5.1-56.8, p<0.001] after adjusting age, sex and the usage of anti-TNF agents.

Then we conducted subgroup analysis on 965 patients with baseline QFT-G (-) and follow-up QFT-G tests. Seroconversion was documented in 65 patients (6.7%). Active tuberculosis was observed in 4 patients, and seroconversion was occurred before the development of active tuberculosis in all patients. The incidence of active tuberculosis in seroconversion patients were 10.5/1000PY.

Conclusion:

QFT-G (+) and QFT-G seroconversion is associated with increased risk of the development of active tuberculosis in patients with AS.

Figure 1.

Overview of patient flow

Disclosure of Interests:

None declared

THU0429 THE ASSOCIATION OF SERUM VITAMIN A, VITAMIN E, AND FOLATE WITH HYPERURICEMIA: AN ANALYSIS OF POPULATION-BASED NATIONALLY REPRESENTATIVE DATA

Background:

Hyperuricemia is an important risk factor for gout as well as hypertension, type 2 diabetes and renal impairment. Although previous studies investigated the association of questionnaire-based micronutrient intake with serum uric acid levels, limited data on serum micronutrients levels in relation to the risk of hyperuricemia especially in Asian population.

Objectives:

This study aimed to evaluate the association of serum vitamin A, vitamin E and folate level with hyperuricemia in the Korean general population.

Methods:

The present study included 6023 participants (2722 men and 3301 women) aged ≥19 years with available data on serum vitamin A, vitamin E, folate and serum uric acid. General characteristics of participants were compared using the Chi-square test and Student’s t test. The association between serum vitamin A, E and folate and serum uric acid levels were evaluated using general linear regression model. Multivariate logistic regression analyses were performed to estimate the effects of these micronutrients on hyperuricemia.

Results:

Serum uric acid levels were increased from the lowest quintile of vitamin A levels to the highest quintile after adjustment for covariates (Ptrend< 0.001 in both sexes). In addition, dose-dependent relationship was observed between vitamin A levels and the risk of hyperuricemia in fully-adjusted analyses (Ptrend< 0.001 in both sexes). However, neither serum vitamin E nor serum folate was associated with hyperuricemia across analyses models.

Conclusion:

This study suggested that vitamin A could be a risk factor of hyperuricemia and further studies are warranted to elucidate underlying mechanism of the observed findings.

References:

[1]Choi, Woo-Joo, et al. “Independent association of serum retinol and β-carotene levels with hyperuricemia: A national population study.”Arthritis care & research64.3 (2012): 389-396.

Disclosure of Interests: :

None declared

0073 Left Anterior Cingulate Hyperarousal During Sleep Anxiety-Inducing Emotional Tasks Performance in Patients with Insomnia Disorder

Introduction

Patients with insomnia frequently experience sleep/insomnia-related anxiety; this anxiety has been associated with hyperarousal. We investigated the underlying brain function changes in patients with insomnia during emotional task performance that induced sleep/insomnia-related anxiety.

Methods

Functional magnetic resonance imaging (fMRI) was performed during emotional task performance in healthy individuals and patients with insomnia who met the diagnostic criteria of insomnia disorder based on the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition, and had chronic insomnia for more than 6 months. The participants underwent fMRI scanning during three types of emotional task performance—insomnia-anxiety task, reading sentences that cause insomnia-related anxiety; general-anxiety task, reading sentences that cause anxiety for everyone; and neutral task, reading neutral sentences that do not cause emotional anxiety. The images obtained from fMRI and blood oxygen level-dependent (BOLD) signal changes were compared between patients with insomnia and healthy controls. Interim analysis was performed with the data of 13 patients with insomnia and 9 controls.

Results

The brain activity in the left anterior cingulate was higher during insomnia-anxiety task performance than that during general-anxiety task performance in the insomnia group (voxel-wise uncorrected p &lt; 0.05; cluster size, 100). In the insomnia group, the brain activity during insomnia-anxiety task performance was not lower in any brain area than that during general-anxiety task performance.

Conclusion

We show that patients with chronic insomnia experience sleep anxiety related with hyperarousal in the left anterior cingulate area. Additional subject recruitment and re-analysis are needed to confirm the findings of this interim analysis.

Support

This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (NRF-2017R1D1A1B03032431).

The effect of platelet-rich plasma as a scaffold in regeneration/revitalization endodontics of immature permanent teeth assessed using 2-dimensional radiographs and cone beam computed tomography: a randomized controlled trial

AIM

To assess the outcomes of platelet-rich plasma as a scaffold in regenerative/revitalization endodontics (RET) using cone beam computed tomography (CBCT) and 2-dimensional radiographs.

METHODOLOGY

Twenty-six healthy patients with mean age of 12.66 ± 4.47, and immature permanent anterior teeth with necrotic pulps, were randomly allocated to two groups, whereby RET was performed using platelet-rich plasma (PRP, test group) and blood clot (BLC, control group). Changes in root length (RL), root dentinal thickness (RDT), apical foramen width (AFW) and radiographic root area (RRA), were assessed using both radiographic methods, whilst changes in periapical area diameter (PAD) were assessed using CBCT, over a period of 12 months. T-test and chi-square/Fisher's exact tests were used to compare continuous and categorical data between BLC and PRP groups, respectively. Changes in RL, RDT, AFW, RRA and PAD were examined by comparing the two groups (PRP versus BLC) using multilevel modelling, considering the clustering effect of repeated measures of several teeth originating from the same participant.

RESULTS

Changes in RL, RDT, AFW, RRA and PAD, over time, were found to be significant for both groups. There was, however, no difference between the RET techniques (PRP versus BLC), using both radiographic and CBCT methods. The results of both assessment techniques (CBCT and 2-dimensional radiographic methods) were highly consistent (overall ICC ranged between 0.80 and 0.94). In addition, a significant effect of baseline PAD was found on RL, RRA and AD at 12 months (RL effect = -0.68, P < 0.001; RRA effect = -1.91, P = 0.025; AD effect = 0.08, P = 0.024).

CONCLUSION

The current study highlights successful and comparable clinical and radiographic outcomes of RET techniques using PRP and BLC. Standardized and calibrated 2-dimensional radiographic assessment was as effective as CBCT in assessing RET outcomes; therefore, the routine use of CBCT in RET is not recommended. Although an effect of baseline periapical lesion diameter on root development outcomes, at 12 months, were observed, more studies are recommended in order to assess such an effect.