Deep learning-based NT-proBNP prediction from the ECG for risk assessment in the community.

OBJECTIVES: The biomarker N-terminal pro B-type natriuretic peptide (NT-proBNP) has predictive value for identifying individuals at risk for cardiovascular disease (CVD). However, it is not widely used for screening in the general population, potentially due to financial and operational reasons. This study aims to develop a deep-learning model as an efficient means to reliably identify individuals at risk for CVD by predicting serum levels of NT-proBNP from the ECG. METHODS: A deep convolutional neural network was developed using the population-based cohort study Hamburg City Health Study (HCHS, n=8,253, 50.9 % women). External validation was performed in two independent population-based cohorts (SHIP-START, n=3,002, 52.1 % women, and SHIP-TREND, n=3,819, 51.2 % women). Assessment of model performance was conducted using Pearson correlation (R) and area under the receiver operating characteristics curve (AUROC). RESULTS: NT-proBNP was predictable from the ECG (R, 0.566 [HCHS], 0.642 [SHIP-START-0], 0.655 [SHIP-TREND-0]). Across cohorts, predicted NT-proBNP (pNT-proBNP) showed good discriminatory ability for prevalent and incident heart failure (HF) (baseline: AUROC 0.795 [HCHS], 0.816 [SHIP-START-0], 0.783 [SHIP-TREND-0]; first follow-up: 0.669 [SHIP-START-1, 5 years], 0.689 [SHIP-TREND-1, 7.3 years]), comparable to the discriminatory value of measured NT-proBNP. pNT-proBNP also demonstrated comparable results for other incident CVD, including atrial fibrillation, stroke, myocardial infarction, and cardiovascular death. CONCLUSIONS: Deep learning ECG algorithms can predict NT-proBNP concentrations with high diagnostic and predictive value for HF and other major CVD and may be used in the community to identify individuals at risk. Long-standing experience with NT-proBNP can increase acceptance of such deep learning models in clinical practice.

GATA6 mutations in hiPSCs inform mechanisms for maldevelopment of the heart, pancreas, and diaphragm.

Damaging GATA6 variants cause cardiac outflow tract defects, sometimes with pancreatic and diaphragmic malformations. To define molecular mechanisms for these diverse developmental defects, we studied transcriptional and epigenetic responses to GATA6 loss of function (LoF) and missense variants during cardiomyocyte differentiation of isogenic human induced pluripotent stem cells. We show that GATA6 is a pioneer factor in cardiac development, regulating SMYD1 that activates HAND2, and KDR that with HAND2 orchestrates outflow tract formation. LoF variants perturbed cardiac genes and also endoderm lineage genes that direct PDX1 expression and pancreatic development. Remarkably, an exon 4 GATA6 missense variant, highly associated with extra-cardiac malformations, caused ectopic pioneer activities, profoundly diminishing GATA4, FOXA1/2, and PDX1 expression and increasing normal retinoic acid signaling that promotes diaphragm development. These aberrant epigenetic and transcriptional signatures illuminate the molecular mechanisms for cardiovascular malformations, pancreas and diaphragm dysgenesis that arise in patients with distinct GATA6 variants.

Alcohol consumption alters Gdnf promoter methylation and expression in rats.

Alcohol use disorder is one of the most disabling diseases worldwide. Glial-cell derived neurotrophic factor (Gdnf) shows promising results concerning the inhibition of alcohol consumption in rodent models. We investigated the epigenetic regulation of Gdnf following ethanol consumption and withdrawal in a rat model. 32 Wistar rats underwent 7 weeks of intermittent access to alcohol in a 2-bottle choice (IA2BC). Whole blood, Nucleus Accumbens (NAc) and Ventral Tegmental Area (VTA) were collected immediately after the last 24 h of an alcohol-drinking session (alcohol group, AG) or 24 h after withdrawal (withdrawal group, WG). MRNA levels were measured using real-time quantitative PCR. Bisulfite-conversion of DNA and capillary sequencing was used to determine methylation levels of the core promoter (CP) and the negative regulatory element (NRE). The CP of the AG in the NAc was significantly less methylated compared to controls (p < 0.05). In the NAc, mRNA expression was significantly higher in the WG (p < 0.05). In the WG, mRNA expression levels in the VTA were significantly lower (p < 0.05) and showed significantly less methylation in the NRE in the VTA (p < 0.001) and the NAc (p < 0.01) compared to controls. Changes in the cerebral mRNA expression correspond to alterations in DNA methylation of the Gdnf promoter in a rodent model. Our results hold clinical relevance since differences in Gdnf mRNA expression and DNA methylation could be a target for pharmacological interventions.

SarcTrack.

RATIONALE: Human induced pluripotent stem cell-derived cardiomyocytes (hiPSC-CMs) in combination with CRISPR/Cas9 genome editing provide unparalleled opportunities to study cardiac biology and disease. However, sarcomeres, the fundamental units of myocyte contraction, are immature and nonlinear in hiPSC-CMs, which technically challenge accurate functional interrogation of contractile parameters in beating cells. Furthermore, existing analysis methods are relatively low-throughput, indirectly assess contractility, or only assess well-aligned sarcomeres found in mature cardiac tissues. OBJECTIVE: We aimed to develop an analysis platform that directly, rapidly, and automatically tracks sarcomeres in beating cardiomyocytes. The platform should assess sarcomere content, contraction and relaxation parameters, and beat rate. METHODS AND RESULTS: We developed SarcTrack, a MatLab software that monitors fluorescently tagged sarcomeres in hiPSC-CMs. The algorithm determines sarcomere content, sarcomere length, and returns rates of sarcomere contraction and relaxation. By rapid measurement of hundreds of sarcomeres in each hiPSC-CM, SarcTrack provides large data sets for robust statistical analyses of multiple contractile parameters. We validated SarcTrack by analyzing drug-treated hiPSC-CMs, confirming the contractility effects of compounds that directly activate (CK-1827452) or inhibit (MYK-461) myosin molecules or indirectly alter contractility (verapamil and propranolol). SarcTrack analysis of hiPSC-CMs carrying a heterozygous truncation variant in the myosin-binding protein C ( MYBPC3) gene, which causes hypertrophic cardiomyopathy, recapitulated seminal disease phenotypes including cardiac hypercontractility and diminished relaxation, abnormalities that normalized with MYK-461 treatment. CONCLUSIONS: SarcTrack provides a direct and efficient method to quantitatively assess sarcomere function. By improving existing contractility analysis methods and overcoming technical challenges associated with functional evaluation of hiPSC-CMs, SarcTrack enhances translational prospects for sarcomere-regulating therapeutics and accelerates interrogation of human cardiac genetic variants.