Deep Phenotyping and Prediction of Long-term Cardiovascular Disease: Optimized by Machine Learning

BackgroundPrediction of cardiovascular disease (CVD) is important in clinical practice. Machine learning (ML) may offer an improved alternative to current CVD risk stratification in individual patients. We aim to identify important predictors and compare ML models with traditional models according to their prediction performance in a large long-term follow-up cohort.MethodsThe Atherosclerosis Risk in Communities (ARIC) study was designed to study the progression of subclinical disease to cardiovascular events over a 25-year follow-up period. All phenotypic variables at visit 1 were obtained. All-cause death, CVD, and coronary heart disease were the outcomes for analysis. The ML framework involved variable selection using the random survival forest (RSF) method, model building, and 5-fold cross-validation. Model performance was evaluated by discrimination using the Harrell concordance index (C-index), accuracy using the Brier score (BS), and interpretability using the number of variables in the model.ResultsOf the 14,842 participants in ARIC, the average age was 54.2 years, with 45.2% male and 26.2% Black participants. Thirty-eight unique variables were selected in the RSF top 20 importance ranking of all 6 outcomes. Aging, hypertension, glucose metabolism, renal function, coagulation, adiposity, and sodium retention dominated the predictions of all outcomes. The ML models outperformed the regression models and established risk scores with a higher C-index, lower BS, and varied interpretability.ConclusionsThe ML framework is useful for identifying important predictors of CVD and for developing models with robust performance compared with existing risk models.     

The Impact of Testosterone on the QT Interval: A Systematic Review

Humans and mammals have sex-specific differences in cardiac electrophysiology, linked to the action of sex hormones in the cardiac muscle. These hormones can upregulate or downregulate the expression of ionic channels modulating the cardiac cycle through genomic and non-genomic interactions. Systematic search in PubMed, Medline and EMBASE including keywords pertaining to testosterone and QT interval. Included experimental studies and observation studies and case reports presenting the results of testosterone administration, excess or deficiency in humans and animals. Testosterone has been shown to shorten the action potential duration, by enhancing the expression of K⁺ channels and downregulating I_CaL increasing the repolarization reserve of the cardiac muscle. This effect has been observed in both genders and animals. Testosterone deficient states can promote arrhythmogenesis. The evidence in this paper may be used to guide clinical considerations, such as increased clinical surveillance of patients in testosterone deficient states using ECG.     

Progress in Targeting the TGFβ Pathway

Signaling by the transforming growth factor-β (TGF-β) superfamily is important in the regulation of hematopoiesis and is dysregulated in myelodysplastic syndromes (MDS), contributing to ineffective hematopoiesis and clinical cytopenias. TGF-β, activins and growth differentiation factors exert inhibitory effects on red cell formation by activating canonical SMAD2/3 pathway signaling. SMAD2/3 overactivation is seen in numerous subtypes of MDS. Furthermore, reduced levels of inhibitory SMAD7 are

References

• 1.Valcarcel D, Verma A, Platzbecker U, et al. Phase 2 Study of Monotherapy Galunisertib (LY2157299 Monohydrate) in Very Low-, Low-, and Intermediate-Risk Patients with Myelodysplastic Syndromes. Blood. 2015;126:1669.
• 2.Suragani RN, Cawley SM, Li R, Wallner S, et al. Modified activin receptor IIB ligand trap mitigates ineffective erythropoiesis and disease complications in murine β-thalassemia. Blood. 2014 Jun 19;123(25):3864-72.
• 3.Dussiot M, Maciel TT, Fricot A, et al. Nat Med. 2014 Apr;20(4):398-407.
• 4.