Abstract Background Variability in cardiovascular conditions is partly explained by common genomic variation identifiable by genome-wide association studies (GWAS). Genetic substrate may overlap in certain entities, this being quantified by genetic correlation. We aimed to leverage multi-trait analysis of GWAS (MTAG), (1) which boosts statistical power by joint analysis of genetically correlated traits, to improve genetic loci discovery in 3 common cardiovascular diseases: atrial fibrillation (AF), coronary artery disease (CAD), and heart failure (HF). Methods and Results Cardiovascular phenotypes with available GWAS summary statistics most correlated with the disease of interest were analysed jointly by MTAG (workflow depicted in Figure 1). These included body mass index and HF for AF (2); HF, LDL-cholesterol, and systolic blood pressure for CAD (3); and AF, CAD, and dilated cardiomyopathy for HF (4). MTAG increased statistical power for all 3 disease GWAS and allowed the identification of 19 new genomic loci for AF, 89 for CAD and 52 for HF (at a GWAS significance level P<5x10-8) (Figure 2). New loci were defined as not previously identified in the single trait original GWAS. Consistently, 19/19, 85/89 and 48/52 of new loci reached nominal significance (P<0.05) in the original studies; and 37%, 44% and 36% of new loci have been associated with AF, CAD and HF in GWAS and phenome wide association studies (PheWAS) in independent cohorts. MTAG summary statistics annotation and functional analysis integrated the distance to the canonical transcriptional start site, chromatin interaction experiments, quantitative molecular phenotype trait experiments and in silico functional prediction through OpenTargets Genetics. MAGMA gene-set functional analysis of the new MTAG results showed 87, 27 and 24 significant associated pathways to AF, CAD and HF respectively. Among the top associations, cardiac depolarization, repolarization, and muscle contraction in AF; macromolecular complex remodeling, triglyceride, chylomicron and LDL metabolism in CAD; and muscle structure, tissue development and RNA transcription initiation in HF. Among novel relevant genes in AF, SRR on chromosome 16 is involved in central nervous system development through regulation of NMDA receptors. This locus has been previously associated with QRS in another GWAS and SRR is related to reduction in heart rate variability and AF enhancement in rats. In CAD, RSPO3 and JAG1 genes, on chromosomes 6 and 20 respectively, have a relevant role in angiogenesis, where RSPO3 is a crucial regulator of coronary artery formation in the developing heart. Conclusions MTAG allows the discovery of new trait-specific loci involved in the pathogenesis of AF, CAD, and HF congruent with the mechanisms of each disease and consistent with independent external cohort PheWAS and GWAS. Thus, MTAG is an interesting tool to unravel relevant mechanisms and potential therapeutic targets in common cardiovascular diseases.

Hypertrophic cardiomyopathy (HCM) is an important cause of morbidity and mortality with both monogenic and polygenic components. We here report results from the largest HCM genome-wide association study (GWAS) and multi-trait analysis (MTAG) including 5,900 HCM cases, 68,359 controls, and 36,083 UK Biobank (UKB) participants with cardiac magnetic resonance (CMR) imaging. We identified a total of 70 loci (50 novel) associated with HCM, and 62 loci (32 novel) as sociated with relevant left ventricular (LV) structural or functional traits. Amongst the common variant HCM loci, we identify a novel HCM disease gene,SVIL, which encodes the actin-binding protein supervillin, showing that rare truncatingSVILvariants cause HCM. Mendelian randomization analyses support a causal role of increased LV contractility in both obstructive and non-obstructive forms of HCM, suggesting common disease mechanisms and anticipating shared response to therapy. Taken together, the findings significantly increase our understanding of the genetic basis and molecular mechanisms of HCM, with potential implications for disease management.

Abstract Funding Acknowledgements Type of funding sources: Public grant(s) – EU funding. Main funding source(s): Horizon 2020 European Commission Project H2020-MSCA-ITN-2016 (764738) and the Clinical Research in Cardiology grant from the Spanish Cardiac Society. Background The aetiology of left ventricular hypertrophy (LVH) is a relevant clinical challenge with consequences for patient management. Phenotypes resulting from hypertensive remodelling and sarcomere mutation often overlap. Synchrotron X-ray phase-contrast imaging (X-PCI) is a technique that can provide 3-dimensional detailed information on myocardial micro-structure non-destructively. The aim is to relate macrostructural/functional, non-invasive, imaging phenotypes of hypertrophic cardiomyopathy (HCM) to the underlying myocardial microstructure assessed with X-PCI. Methods Myocardial tissue samples were obtained from three patients (P1-3) with obstructive myocardial hypertrophy undergoing septal myectomy. Medical history and the 5-year HCM risk scores were evaluated. The patients were imaged with magnetic resonance imaging and echocardiography prior to procedure. Myocardial structure was assessed with wall thickness, late gadolinium enhancement (LGE), whereas function with speckle-tracking deformation (STE) and tissue Doppler imaging (TDI). Myectomy tissue was imaged with X-PCI in the TOMCAT beamline, using a multiscale propagation-based protocol combining a low-resolution (LR) and a high-resolution (HR) setup (5.8 and 0.7 um pixel size, respectively). Results The clinical and imaging data are shown in Fig 1. On initial assessment, wall thickness, LGE distribution, global longitudinal strain and septal TDI demonstrated a similar macrostructural and functional phenotype of P1 and P2, whereas P3 stood out with more severe hypertrophy, scarring and dysfunction. Additional regional deformation analysis with STE revealed reduced deformation in the basal and mid septum in P1, paired with a hypertensive pattern of post-systolic shortening (PSS) (yellow arrows). In comparison, in P2 and P3, deformation was more heterogeneous regionally, with regions of almost complete absence of deformation (orange arrows). Upon further exploration with TDI, areas with abnormal deformation were identified on the transition from basal to mid septum in both P2 and P3, whereas deformation was normal, but reduced in P1, and paired with PSS. LR X-PCI defined regions of interest to scan with HR (yellow frame), where HR revealed extensive interstitial fibrosis (orange arrow) with normal myocyte size and organisation in P1, compatible with severe hypertensive remodelling. However, in P2 and P3, patches of fibrosis (yellow arrow) paired with enlarged myocytes organized in visible disarray, considerably more prominent in P3, were both compatible with sarcomere-mutation HCM. Conclusion The results demonstrate multiscale phenotyping of HCM - relating micro- and macrostructural findings to function, and integrating multimodality data. In-depth regional deformation analysis, validated by synchrotron-based microstructural analysis, showed potential to identify distinct imaging phenotypes in HCM, distinguishing between overlapping presentations in different aetiologies. Abstract Figure 1

The number of patients with Chagas disease residing in Europe has increased significantly due to migration flows. Globally, Chagas cardiomyopathy has worse prognosis than other types of dilated cardiomyopathies and about 30% of patients develop cardiac involvement after a variable latency period (10–30 years). However, there is lack of data regarding the evolution of patients with Chagas disease living in a non-endemic area and potential early predictors of disease progression. OBJETIVE To describe the natural course of Chagas disease, the incidence rate of transformation into cardiac form and to assess if early predictors of myocardial involvement translate into a worse long-term prognosis in our non-endemic cohort. Clinical and echocardiographic follow-up was performed in 202 individuals from endemic areas of Chagas disease. At baseline, electrocardiogram, BNP and a comprehensive echocardiography including diastolic function and longitudinal myocardial strain were performed. Four different groups were defined: healthy controls (N=77); Chagas indeterminate form (positive serology, normal ECG and left ventricle (LV) dimensions and LV ejection fraction (>50%) and no segmental abnormalities, N=92); Chagas patients with abnormal ECG but normal LV dimensions and motility (N=15); and Chagas patients with LV diameter>55 mm or LV ejection fraction<50% or segmental abnormalities (N=18). The primary clinical outcome included advanced atrioventricular block, sustained ventricular tachycardia, heart failure, heart transplant, death or progression of cardiac disease defined as LV systolic dysfunction or new segmental abnormalities. Kaplan Meier with Long rank test and Cox regression analysis was used. Mean age was 37±9 and 34% were male. Median follow-up was 69 months (range 1 to 147). The primary endpoint occurred in a total of 17 (8.4%) individuals: 5 (5.4%) in the Indeterminate group; 3 (20%) in the abnormal ECG group; and 9 (50%) in the group with abnormal LV dimension or motility, with no events among controls (long-rank test<0.01, Figure 2). Six patients evolved from the indeterminate phase to cardiac involvement (2 with isolated ECG changes and 4 with abnormal echocardiography without previous changes in ECG (Figure 1). On echocardiography, there were no differences regarding changes in LV dimensions or LV ejection fraction between Chagas patients with normal baseline echo and controls, but a significantly reduction of Em was observed (−1.6±3.0 vs. 0.2±1.0) in the former. Excluding patients with abnormal echo at baseline, BNP (HR=1.03, p=0.001), Em (HR=0.78, p=0.05) and left atrial diameter (1.23, p=0.01) were predictors of the combined event. Conversion from the indeterminate to Chagas cardiomyopathy in our cohort was approximately 1.1%/year, but it may happen directly with contractility disturbances. BNP and comprehensive echocardiography may help to early detect disease progression. Figure 1. Distribution of patients and KM curves Type of funding source: None

Abstract BACKGROUND Pharmacologically induced pulmonary hypertension (PH) is infrequent nowadays and it is included in the type 1 of the classification of PH. Tyrosine kinase inhibitors (TKI) are the cornerstone of the treatment of many haemotopoietic stem cell diseases. Dasatinib is a second-generation TKI used in chronic myeloid leukemia (CML) and as an infrequent cardiovascular side-effect (< 0,50%) could induce PH, usually reversible but life threatening. Only a few case series are published. CASE DESCRIPTION: We present a 51-year-old woman who was diagnosed of a CML when she was 46. Initially, she underwent therapy with imatinib but after 5 years of treatment she developed resistance to this drug, and dasatinib was prescribed as a second line drug. After 3 months of continuous treatment, she started with dry cough and effort dyspnea. Blood analysis, EKG and Chest X-Ray were made but did not show outstanding findings. An unspecific viral infection was the final diagnosis. The patient clinical condition deteriorated with major dyspnea and edemas in the lower limbs. A TTE showed moderate tricuspid regurgitation and severe HP systolic pulmonary artery pressure (sPAP) of 80 mmHg. The pulmonary acceleration time was shortened and a mesosistolic knock was present. Systolic dysfunction of the right ventricle and pericardial effusion (image 1,2,3,4) were noted. The right atrium was not dilated. Cava vein was dilated but with inspiratory collapse >50%. The left ventricular function was preserved, but first degree diastolic dysfunction was found. Other causes of PH were excluded (types 2, 3, 4). A CT pulmonary angiogram did not show segmental perfusion defects. Finally, a right heart catheterization confirmed the TTE findings: severe precapillary PH without postcapillary component. After the diagnosis was confirmed, TKI was stopped and double targeted therapy with ambrisentan + tadalafil was started. After 6 months of treatment a new TTE was made with complete reversal of the secondary changes in the myocardium induced by the PH. No tricuspid regurgitation was detected nor any indirect sign of PH was found. (image 5,6). CONCLUSIONS Drug-induced PH is rare nowadays and most cases were described in the seventies in the USA related with the epidemic of anorexigenic drugs. Although the pathogenesis still remains unclear, treatment includes immediately stopping the offending agent. Echocardiography due to its accessibility, reproducibility, consistence and low cost should be the first diagnostic tool to be considered, because as it is known, in the early stages of the disease, before developing right disfunction, clinical and conventional tests are non-specific. Abstract P1721 Figure. Echo images: previous and afte treatment

Abstract Background Brain natriuretic peptide (BNP) and novel echocardiographic techniques such as speckle-tracking and a comprehensive evaluation of diastolic function can detect early myocardial involvement in patients with Chagas disease. However, there is lack of longitudinal studies that can confirm whether this early myocardial involvement translates into a worse prognosis. Purpose To assess if early myocardial involvement detected by BNP or a comprehensive echocardiographic evaluation was associated with future events in Chagas disease. Methods 182 consecutive individuals from endemic areas who underwent T. cruzi screening where prospectively included from 2007 to 2014. ECG, BNP and a comprehensive echocardiography including diastolic function and longitudinal myocardial strain were performed. Four different groups were defined: healthy controls (N=77); Chagas indeterminate form (positive serology, normal ECG and left ventricle (LV) diameter (<55 mm), LV ejection fraction (>50%) and no segmental abnormalities, N=88); Chagas patients with abnormal ECG but normal LV dimensions and motility (N=7); and Chagas patients with LV diameter>55 mm or LV ejection fraction<50% or segmental abnormalities (N=13). The primary outcome included advanced atrioventricular block, sustained ventricular tachycardia, heart failure, heart transplant or death. Kaplan Meier with Long rank test and Cox regression analysis was used. Results Mean age was 37±9 and 34% were male. Median follow-up was 63 months (range 1 to 137). The primary endpoint occurred in a total of 11 (10%) individuals: 2 (2.4%) in the Indeterminate group; 3 (43%) in the abnormal ECG group; and 6 (46%) in the group with abnormal LV dimension or motility, with no events among controls (long-rank test<0.01, Figure). In the global population, age, BNP, diastolic dysfunction parameters and longitudinal strain at the inferior and lateral walls were significant predictors. In the cohort of Chagas patients with normal standard echocardiography (N=92), ECG abnormalities (HR=49, p=0.001), Em (HR=0.68, p=0.03), deceleration time (HR=0.01, p=0.01), left atrial diameter (HR=1.24, p<0.01) and longitudinal strain at the midventricular lateral wall (HR=0.75, p=0.028) remained significantly associated with outcome. Cumulative survival without events Conclusions Outcome was significantly more frequent in Chagas patients with abnormalities in ECG or standard echocardiography. In early forms of the disease, myocardial involvement detected by BNP or a comprehensive echocardiography was associated with prognosis, and may help to individualize treatment and follow-up.

Abstract Background In chronic pulmonary hypertension (PH), the main cause of death is right ventricular (RV) failure. However, the onset of RV dysfunction varies significantly among patients. Early recognition of RV maladaptation would be highly relevant. Purpose To identify cardiac magnetic resonance (CMR)-derived parameters affected in early stages of PH before the development of RV dysfunction. Methods Experimental chronic PH was generated by pulmonary vein banding in 76 pigs whereas 14 animals underwent sham operation. Animals were followed-up monthly with right heart catheterization (RHC) and immediate CMR for a maximum of 9 months. CMR exams included a T1-mapping sequence to quantify equilibrium-extracellular volume (ECV) at the RV insertion points. Pairs of RHC and CMR examinations were compared among controls and PH with normal RV ejection fraction (RVEF≥55%) or reduced (REVF<55%) using ANCOVA test with Bonferroni correction. All procedures followed the “Principles of laboratory animal care”. Results In the presence of PH confirmed by RHC and normal RVEF, the RV displays significant hypertrophy and increased myocardial native T1 and ECV, despite preserved ventricular dimensions and PA flow (Fig 1B). In advanced stages of the disease, RV-AP uncoupling and reduced PA pulsatility develop together with RV dilatation and failure (Fig 1C). CMR parameters in sham and PH animals Sham controls (n=25 evaluations) PH with normal RVEF (n=155 evaluations) PH with low RVEF (n=70) RVEF (%) 62±5 61±4 47±8# RVEDV (mL/m2) 73±18 83±17 111±29# RVESV (mL/m2) 28±8 32±8 61±25# RV mass (g/m2) 17±4 23±5* 30±11# PA area (cm2/m2) 5.5±1.2 6.3±1.6 8.7±2.2# Native T1 anterior RVIP (ms) 983±75 1043±78* 1055±90# ECV anterior RVIP (%) 27±5 31±6* 36±7# Native T1 inferior RVIP (ms) 959±68 1022±71* 1032±99# ECV inferior RVIP (%) 25±5 31±6* 36±7# PA pulsatility (%) 29±6 27±7 21±3# Ea/Emax 40±16 47±11 97±65# RVEDV: RV end-diastolic volume; RVESV: RV end-systolic volume; RVIP: RV insertion point. *p<0.05 1 vs. 2; #1 vs. 3. NativeT1: control, PH-normal RV, RV dysf Conclusion RV hypertrophy and ECV expansion are early mechanisms in RV adaptation to postcapillary PH, whereas ventricular and PA dilatation, RV-PA uncoupling and reduced pulsatility appear in more advanced stages concurring with systolic dysfunction. Acknowledgement/Funding The CNIC is supported by the Ministerio de Ciencia, Innovaciόn y Universidades and the Pro CNIC Foundation, and is a Severo Ochoa Center of Excellence

Abstract Background Right ventricular (RV) dysfunction is the most important prognostic factor in chronic pulmonary hypertension (PH), but its underlying mechanisms are unknown. Clinical observation and prior experimental work suggest that RV pressure overload is not the only cause since the degree of RV adaptation varies with similar RV end-systolic pressures. Purpose Our aim was to characterize serial RV adaptation by cardiac magnetic resonance (CMR) in 3 different experimental large-animal models of increased afterload: a model of chronic postcapillary PH, a model of PH secondary to systemic-to-pulmonary shunt and a model of mechanical RV pressure overload (generated by pulmonary artery [PA] banding). Methods Four-week old piglets underwent pulmonary vein banding surgery to generate the chronic postcapillary PH model (n=20), aorto-pulmonary shunt (n=6), PA banding (n=7) or sham operation (n=7). They were followed up monthly with CMR and right heart catheterization (RHC). All procedures followed the “Principles of laboratory animal care”. Comparison of continuous variables among groups was performed with Mann-Whitney U test. Results Animals with either postcapillary PH or PH secondary to aorto-pulmonary shunt presented significant RV dilatation, hypertrophy and dysfunction that was maintained during follow-up (median RV end-systolic volume [RVESV]=32.6 ml/m2 for postcapillary PH and 32.6 ml/m2 for shunt vs. 16.1 ml/m2 in sham controls; median RV ejection fraction [RVEF]=61.5% for postcapillary PH and 60.5% for shunt vs. 69.6% in sham controls at the end of follow-up). Animals with PA banding also presented with significant RV dilatation and hypertrophy at the first month follow-up, but unlike all other groups, they developed reverse RV remodeling from the second month onwards and maintained normal RV volumes and RVEF values until the end of follow-up despite having severe RV hypertrophy (RV mass 22.6 g/m2 in PA banding vs. 16.1 g/m2 in controls at the 4th month follow-up; Figure). CMR parameters (median values). Conclusion In PH there is a maladaptive RV hypertrophy that is not present in a model of progressive RV pressure overload without alterations of the pulmonary circulation. Increased RV pressure overload alone does not fully explain PH-associated RV dysfunction. Further research is needed to clarify the underlying mechanisms of adaptive and maladaptive hypertrophy in PH. Acknowledgement/Funding The CNIC is supported by the Ministerio de Ciencia, Innovaciόn y Universidades and the Pro CNIC Foundation, and is a Severo Ochoa Center of Excellence