Derivation and validation of a machine learning-based risk prediction model for in-hospital mortality in patients with acute heart failure

Background

Although risk stratification is important in patients with acute heart failure (AHF) to predict patient prognosis, pre-existing risk models have not often been used due to its complexity. Recently, machine learning methods have been presented as an alternative approach to analyzing the predictive probability of large clinical datasets.

Purpose

The aim of this study is to develop a user-friendly risk score developed by one of machine learning methods and compare the performance of the new risk score to the existing conventional risk models.

Methods

A machine-learning-based risk model was developed using least absolute shrinkage and selection operator (LASSO) regression by identifying predictors of in-hospital mortality in the derivation cohort (REALITY-AHF) and externally validating and comparing its performance with two pre-existing risk models: the Get With The Guidelines risk score incorporating brain natriuretic peptide and hypochloremia (GWTG-BNP-Cl-RS) and the acute decompensated heart failure national registry (ADHERE) risk model.

Results

In-hospital deaths in the derivation and validation (NARA-HF) cohorts were 76 (5.1%) and 61 (4.9%), respectively. The risk score comprised four variables (systolic blood pressure, blood urea nitrogen, serum chloride, and C-reactive protein) and was developed according to the results of the LASSO regression weighting the coefficient for selected variables using a logistic regression model (4V-RS). Even though 4V-RS comprised fewer variables, In the validation cohort, it showed a higher area under the receiver operating characteristic curve (AUC) than the ADHERE risk model (AUC, 0.783 vs. 0.740; P=0.059) and a significant improvement in net reclassification (0.359; 95% CI, 0.10–0.67; p=0.006). 4V-RS performed similarly to GWTG-BNP-Cl-RS in terms of discrimination (AUC, 0.783 vs. 0.759; p=0.426) and net reclassification (0.176; 95% CI, −0.08–0.43; p=0.178).

Conclusions

The 4V-RS model comprising only four readily available data points at the time of admission performed similarly to the more complex pre-existing risk model in patients with AHF.

Funding Acknowledgement

Type of funding sources: Foundation. Main funding source(s): Cardiovascular Research Fund

55Identification of celastramycin as a novel therapeutic agent for pulmonary arterial hypertension - high-throughput screening of 5,562 compounds

Background

Pulmonary arterial hypertension (PAH) is characterized by enhanced proliferation of pulmonary artery smooth muscle cells (PASMCs) accompanying increased production of inflammatory factors and adaptation of mitochondrial metabolism to a hyperproliferative state. However, at present, since all the drugs in clinical use target pulmonary vascular dilatation, they may not be so effective for patients with advanced PAH.

Purposes

We aimed to discover a novel drug for PAH that inhibits PASMC proliferation.

Methods

In the first screening, we examined 5,562 compounds from our original library using high-throughput screening system to discover a compound that inhibits proliferation of PASMCs from PAH patients (PAH-PASMCs). In the second screening, we performed concentration-dependent assays and counter assays with PAH-PASMCs and PASMCs from healthy donors. We also performed apoptosis assays and mechanistic analysis for inflammation, reactive oxygen species (ROS), and mitochondrial function.

Results

We found that celastramycin, a benzoyl pyrrole-type compound originally found in a bacteria extract, inhibited the proliferation of PAH-PASMCs in a dose-dependent manner with minimal effects on PASMCs from healthy donors. Moreover, celastramycin inhibited proliferation with minimal increase in apoptosis and low rate of cell death. Then, we synthesized 25 analogues of celastramycin, and finally selected the lead compound that significantly inhibited proliferation of PAH-PASMCs and reduced cytosolic ROS levels. Mechanistic analysis demonstrated that celastramycin reduced the protein levels of hypoxia-inducible factor-1α, which was abnormally activated in PAH-PASMCs and impaired aerobic metabolism, and nuclear factor-κB, which induces pro-inflammatory signals, in PAH-PASMCs compared with vehicle controls, leading to reduced secretion of inflammatory cytokines. Importantly, celastramycin treatment reduced the ROS levels in PAH-PASMCs with increased protein levels of NF-E2-related factor 2 (Nrf2), a master regulator of cellular response against oxidative stress. Furthermore, celastramycin treatment improved mitochondrial energy metabolism with recovered mitochondrial network formation in PAH-PASMCs. We also discovered that celastramycin-mediated effects on these transcriptional modulators could be regulated by zinc finger C3H1 domain-containing protein, which is a binding partner of celastramycin. Finally, celastramycin treatment ameliorated pulmonary hypertension in three experimental animal models of PH in mice and rats, accompanied by reduced inflammatory changes in the lungs.

Conclusions

These results indicate that celastramycin ameliorates pulmonary hypertension through inhibition of excessive proliferation of PAH-PASMCs, for which its anti-inflammatory and beneficial effects on mitochondrial energy metabolism may be involved. Thus, celastramycin could be a novel drug for PAH as it exerts anti-proliferative effects on PAH-PASMCs.

P6479Diagnostic and prognostic significance of serum levels of selenoprotein P in patients with pulmonary arterial hypertension

Background

Despite the recent progress in upfront combination therapy for pulmonary arterial hypertension (PAH), a useful biomarker for the disorder still remains to be developed. Selenoprotein P (SeP) is a glycoprotein secreted mainly from hepatocytes but also from other various kinds of cells, including pulmonary artery smooth muscle cells (PASMCs), to maintain selenium homeostasis and cellular energy metabolism. We have recently demonstrated that SeP expression in PASMCs is markedly up-regulated in PAH patients and plays crucial roles in the pathogenesis of the disorder. In this study, we thus examined whether serum levels of SeP could be a useful biomarker for the disorder.

Methods

In the experimental study, we performed gene expression microarray and in silico analyses to identify a novel therapeutic target for PAH. We also used the lung, serum, and cultured PAMSCs derived from patients with PAH for mechanistic experiments. In the clinical study, we enrolled a total of 65 consecutive patients with PAH who underwent right heart catheterization for hemodynamic assessment. We measured serum SeP levels and evaluated their prognostic impacts during follow-up (mean 1,520 days, IQR: 1,393–1,804 days). Serum SeP level was measured using a newly developed sol particle homogeneous immunoassay. As controls, we collected serum samples from 20 controls without any known cardiac disorders evaluated by hematological examination, echocardiography, and coronary angiography. In PAH patients, we examined the relationship between baseline SeP levels and composite endpoint of all-cause death and lung transplantation. The correlation between the absolute changes in SeP and those in hemodynamic parameters during follow-up were also examined.

Results

In the experimental study, SeP promoted PASMC proliferation through increased oxidative stress and mitochondrial metabolic dysfunction, which were associated with activated HIF-1α and dysregulated glutathione metabolism. In the clinical study, PAH patients showed significantly higher levels of serum SeP compared with controls (3.07±0.57 vs. 2.43±0.25 mg/L, P<0.0001). Higher SeP levels (cut-off point, 3.47 mg/L) were significantly associated with the composite endpoint of all-cause death and lung transplantation in PAH patients [HR: 4.85 (1.42 to 16.6), P<0.01]. Importantly, we found that absolute changes in SeP levels in PAH patients significantly correlated with those in mean pulmonary artery pressure, pulmonary vascular resistance, and cardiac index in response to PAH-specific therapy (R=0.78, 0.76, and −0.71, respectively, all P<0.0001). Furthermore, the increases in SeP levels during follow-up predicted the poor outcome in PAH patients [Figure, HR: 4.29 (1.27 to 14.4), P<0.05].

Figure 1

Conclusions

These results indicate that SeP is a novel therapeutic target of PAH and that serum SeP levels are a novel biomarker for diagnosis and assessment of treatment efficacy and long-term prognosis in PAH patients.

Acknowledgement/Funding

Grants-in-aid for Scientific Research from the Japan Agency for Medical Research and Development, Tokyo, Japan (16ek0109176h0001, 17ek0109227h0001).

Measurement and comparison of individual external doses of high-school students living in Japan, France, Poland and Belarus-the 'D-shuttle' project

Twelve high schools in Japan (of which six are in Fukushima Prefecture), four in France, eight in Poland and two in Belarus cooperated in the measurement and comparison of individual external doses in 2014. In total 216 high-school students and teachers participated in the study. Each participant wore an electronic personal dosimeter 'D-shuttle' for two weeks, and kept a journal of his/her whereabouts and activities. The distributions of annual external doses estimated for each region overlap with each other, demonstrating that the personal external individual doses in locations where residence is currently allowed in Fukushima Prefecture and in Belarus are well within the range of estimated annual doses due to the terrestrial background radiation level of other regions/countries.

The value of the Cytobrush for obtaining cells from the uterine cervix

A study was undertaken to evaluate the efficacy of the Cytobrush compared with the cotton-tipped applicator for obtaining cells from the uterine cervix. Four consecutive Papanicolaou smears were obtained in 1,000 patients during a single clinic visit. In every patient, a total of four smears on two slides were submitted: one slide contained endocervical material obtained with a cotton-tipped applicator and with the Cytobrush and the other slide contained ectocervical material obtained with both a cotton-tipped applicator and the Cytobrush. In endocervical smears, the Cytobrush technique produced numerous groups and clusters of cervical columnar cells in contrast to the use of the cotton swab. This was particularly true in older patients and pregnant women. In ectocervical smears, the detection of the squamous component was also higher in the Cytobrush smear than in cotton swab smears. However, the cell yield was not influenced by the cell sampling sequence in either endocervical or ectocervical specimens. It is the conclusion of this study that the utility of the Cytobrush technique for collecting specimens for endocervical and ectocervical smears is greater and provides a higher yield of cells than the cotton swab technique. Wider use of this simple procedure is encouraged.