Technical advances in artificial intelligence (AI) in cardiac imaging are rapidly improving the reproducibility of this approach and the possibility to reduce time necessary to generate a report.

Anthracyclines are anti-neoplastic drugs presenting cardiotoxicity as a side effect. Cardiac troponins (cTn) and echocardiography are currently used to assess cardiac damage and dysfunction, but early biomarkers identifying patients in need of preventive treatments remain a partially met need. Circulating microRNAs (miRNAs) represent good candidates, so we investigated their possible roles as predictors of troponin elevation upon anthracycline treatment. Eighty-eight female breast cancer patients administered with doxorubicin (DOX) or epirubicin (EPI) were divided into four groups basing on drug type and cTn positive (cTn+) or negative (cTn−) levels: DOX cTn−, DOX cTn+, EPI cTn− and EPI cTn+. Blood was collected at baseline, during treatment, and at follow-up. We identified plasma miRNAs of interest by OpenArray screening and single assay validation. Our results showed miR-122-5p, miR-499a-5p and miR-885-5p dysregulation in DOX patients at T0, identifying a signature separating, with good accuracy, DOX cTn− from DOX cTn+. No miRNAs showed differential expression in EPI subjects. Conversely, an anthracycline-mediated modulation (regardless of cTn) was observed for miR-34a-5p, -122-5p and -885-5p. Our study indicates specific circulating miRNAs as possible prediction markers for cardiac troponin perturbation upon anthracycline treatment. Indeed, our findings hint at the possible future use of plasma miRNAs to predict the cardiac responsiveness of patients to different anticancer agents.

Background: Feature selection is a crucial step in machine learning analysis. Currently, many feature selection approaches do not ensure satisfying results, in terms of accuracy and computational time, when the amount of data is huge, such as in ‘Omics’ datasets. Results: Here, we propose an innovative implementation of a genetic algorithm, called GARS, for fast and accurate identification of informative features in multi-class and high-dimensional datasets. In all simulations, GARS outperformed two standard filter-based and two ‘wrapper’ and one embedded’ selection methods, showing high classification accuracies in a reasonable computational time. Conclusions: GARS proved to be a suitable tool for performing feature selection on high-dimensional data. Therefore, GARS could be adopted when standard feature selection approaches do not provide satisfactory results or when there is a huge amount of data to be analyzed.

Arrhythmogenic cardiomyopathy (ACM) is hallmarked by ventricular fibro‐adipogenic alterations, contributing to cardiac dysfunctions and arrhythmias. Although genetically determined (e.g., PKP2 mutations), ACM phenotypes are highly variable. More data on phenotype modulators, clinical prognosticators, and etiological therapies are awaited. We hypothesized that oxidized low‐density lipoprotein (oxLDL)‐dependent activation of PPARγ, a recognized effector of ACM adipogenesis, contributes to disease pathogenesis. ACM patients showing high plasma concentration of oxLDL display severe clinical phenotypes in terms of fat infiltration, ventricular dysfunction, and major arrhythmic event risk. In ACM patient‐derived cardiac cells, we demonstrated that oxLDLs are major cofactors of adipogenesis. Mechanistically, the increased lipid accumulation is mediated by oxLDL cell internalization through CD36, ultimately resulting in PPARγ upregulation. By boosting oxLDL in a Pkp2 heterozygous knock‐out mice through high‐fat diet feeding, we confirmed in vivo the oxidized lipid dependency of cardiac adipogenesis and right ventricle systolic impairment, which are counteracted by atorvastatin treatment. The modulatory role of oxidized lipids on ACM adipogenesis, demonstrated at cellular, mouse, and patient levels, represents a novel risk stratification tool and a target for ACM pharmacological strategies.

(1)Background: Chronic heart failure (CHF) contributes to the overall burden of cardiovascular disease. Early identification of at-risk individuals may facilitate the targeting of precision therapies. Plasma microRNAs are promising circulating biomarkers for their implications with cardiac pathologies. In this pilot study, we investigate the possible exploitability of circulating micro-RNAs (miRNAs) to track chronic heart failure (CHF) occurrence, and progression from NYHA class I to IV. (2)Methods: We screened 367 microRNAs using TaqMan microRNA Arrays in plasma samples from healthy controls (HC) and CHF NYHA-class I-to-IV patients (5/group). Validation was performed by singleplex assays on 10 HC and 61 CHF subjects. Differences in the expression of validated microRNAs were evaluated through analysis of covariance (ANCOVA). Associations between N-terminal pro-BNP (NT-proBNP), left ventricular end-diastolic volume (LVEDV) or peak oxygen uptake (VO2 peak) and plasma microRNA were assessed by multivariable linear regression analysis. (3)Results: Twelve microRNAs showed higher expression in CHF patients vs. HC. Seven microRNAs were associated with NT-proBNP concentration; of these, miR-423-5p was also an independent predictor of LVEDV. Moreover, miR-499-5p was a predictor of the VO2 peak. Finally, a cluster of 5 miRNAs discriminated New York Heart Association (NYHA) class-I from HC subjects. (4)Conclusions: Our data suggest that circulating miRNAs have the potential to serve as pathophysiology-based markers of HF status and progression, and as indicators of pre-symptomatic individuals.