O HDL-c é um fator de risco cardiovascular negativo e sua concentração plasmática apresenta relação inversa com a incidência de eventos cardiovasculares. Entretanto, as evidências relativas ao grupo de indivíduos com níveis de HDL-c acima do percentil 95 da população geral ainda são escassas e o impacto da hiperalfalipoproteinemia (HALP) sobre o risco cardiovascular continua representando motivo de controvérsia na literatura médica. Alguns estudos em populações específicas associam a HALP a aumento do risco cardiovascular. Ao mesmo tempo, outros estudos identificaram populações de indivíduos hipoalfalipoproteinêmicos com marcada longevidade. Assim, demonstrou-se aparente dissociação entre níveis de HDL-c e risco cardiovascular em determinadas populações, reconduzível a aspectos disfuncionais da HDL. O objetivo do presente estudo foi verificar o papel da HALP na determinação do risco cardiovascular; comparar a prevalência de doença cardiovascular subclínica, avaliada por meio da quantificação ultrassonográfica da Espessura Íntimo-Medial Carotídea (EIMC), entre portadores de HDL-c >= 90mg/dL (grupo HALP) e portadores de concentrações de HDL-c atualmente consideradas normais (entre 40 e 50mg/dL para os homens e entre 50 e 60mg/dL para as mulheres); e avaliar características e função da HDL em portadores de HALP por meio do estudo de sua composição, de sua capacidade de efluxo de colesterol, e de sua atividade anti-inflamatória e antioxidante, correlacionando estas características com a presença de doença cardiovascular subclínica avaliada por meio da determinação da EIMC, da Velocidade de Onda de Pulso (VOP) e da presença de Calcificação Arterial Coronariana (CAC) avaliada pela TCMD. Para responder estas perguntas, o presente estudo foi articulado em dois braços: Braço 1: Análise da coorte do estudo ELSA com o objetivo de determinar a prevalência de HALP em uma população geral; definir o perfil demográfico, antropométrico e metabólico dos portadores de HALP; e comparar a prevalência de doença vascular subclínica deste grupo com controles da mesma coorte com níveis normais de HDL-colesterol. Braço 2: Recrutamento de 80 voluntários hígidos e portadores de HALP para avaliação da correlação entre presença de doença vascular subclínica, e aspectos estruturais e funcionais da HDL. Em seus dois braços, o estudo levou a quatro conclusões principais: 1) Níveis marcadamente elevados de HDL-c estão associados a menor espessura íntimo-medial carotídea quando comparados a níveis de HDL-c considerados normais pelas diretrizes vigentes. Embora portadores do fenótipo HALP apresentem, como grupo, um perfil metabólico mais favorável que o encontrado em indivíduos com HDL-c normal, a associação entre EIMC e HALP foi independente dos fatores de risco tradicionais, indicando que a menor prevalência destes últimos em portadores de HDL-c marcadamente elevado justifica apenas parcialmente a menor prevalência de doença vascular subclínica neste grupo; 2) Embora a HALP se apresente como um fenótipo ateroprotetor, há indivíduos com níveis marcadamente elevados de HDL-c que evoluem com doença cardiovascular, clínica ou subclínica. Neste contexto, nossos resultados indicam correlação entre os três métodos avaliados para estudar doença vascular subclínica em portadores de HALP: EIMC, VOP e CAC; 3) Os fatores de risco tradicionais continuam exercendo seu peso na determinação do risco cardiovascular em portadores de HALP. Idade, tabagismo, hipertensão arterial, hipertrigliceridemia e altos níveis de LDL-c apresentaram associação estatisticamente significativa com a presença de doença vascular subclínica no grupo estudado; 4) A avaliação da composição e da função da HDL em portadores de HALP pode permitir identificar indivíduos especificamente mais suscetíveis à aterosclerose. Nossos resultados indicam que, em particular, a atividade anti-inflamatória da HDL, avaliada pela capacidade de inibição da produção de IL-6; o efluxo de colesterol e a capacidade de transferência de triglicérides apresentaram associação independente com menor espessura íntimo-medial carotídea em portadores de HALP, enquanto níveis mais altos de Apo A-IV se associaram a maior grau de doença cardiovascular subclínica / HDL-c is a negative cardiovascular risk factor and its plasma concentration is inversely related to the incidence of cardiovascular events. However, evidence of benefit among subjects with HDL-c levels above the 95th percentile of the general population is still scarce and the impact of hyperalphalipoproteinemia (HALP) on cardiovascular risk continues to represent matter of debate in the medical literature. Some studies with specific populations indicated an increased cardiovascular risk associated with HALP. In addition, other reports identified groups of patients with marked hypoalphalipoproteinemia and longevity. Hence, there could be a dissociation between HDL-c levels and cardiovascular risk in certain populations, possibly due to dysfunctional HDL particles. The aim of this study was to investigate the role of HALP phenotype in determining cardiovascular risk; to compare the prevalence of subclinical cardiovascular disease, assessed by ultrasound measurement of Carotid Intima-Media Thickness (CIMT) among patients with HDL-c >= 90mg/dL (HALP group) and patients with HDL-c currently considered normal (40-50mg/dL for men and 50-60mg/dL for women); and to evaluate HDL functionality in patients with HALP through the study of its composition, its cholesterol efflux capacity, and its anti-inflammatory and antioxidant activity; correlating those characteristics with the presence of subclinical cardiovascular disease assessed by CIMT, Pulse Wave Velocity (PWV) and Coronary Artery Calcification (CAC). To answer these questions, the present study was articulated into two arms: Arm 1: ELSA-Brasil study cohort analysis in order to assess HALP prevalence in a general population, defining demographic, anthropometric and metabolic profile of HALP individuals; and comparing the prevalence of subclinical vascular disease among HALP subjects with controls with normal HDL-c. Arm 2: Recruitment of 80 healthy volunteers with HALP to study the correlation between subclinical vascular disease and HDL functionality in this group. Our study led to four main conclusions: 1) markedly elevated HDL-c is associated with lower CIMT compared to the control group with normal HDL-c levels. Although individuals with HALP display a more favorable metabolic profile than subjects with normal HDL-c, the association between CIMT and HALP was independent of traditional risk factors, indicating that the lower prevalence of subclinical vascular disease in this group is only partially justified by the lower prevalence of cardiovascular risk factors; 2) Although HALP can be regarded as an atheroprotective phenotype, there are individuals with markedly elevated levels of HDL-c who develop cardiovascular disease. Our results indicate good correlation of the three methods here adopted to study subclinical vascular disease among HALP patients: CIMT, VOP and CAC; 3) Traditional risk factors continue to exert their weight in determining cardiovascular risk in patients with HALP. Age, smoking, hypertension, hypertriglyceridemia and high levels of LDL-c were significantly associated with the presence of subclinical vascular disease among HALP individuals; 4) the assessment of the HDL composition and functionality in patients with HALP may allow to identify individuals specifically more susceptible to atherosclerosis. Our results indicate that, in particular, cholesterol efflux capacity, the anti-inflammatory activity of HDL, and triglyceride transfer capacity were independently associated with lower CIMT in HALP individuals, while higher levels of Apo A-IV were associated with a greater burden of subclinical cardiovascular disease

Les maladies complexes telles que les maladies cardio-Vasculaires (MCV) sont influencées par des facteurs génétiques et environnementaux. L’estimation du risque cardio-Vasculaire chez un individu est généralement évaluée par la sommation des facteurs de risque reconnu des MCV (p. ex. l’âge, le sexe, le tabac, la pression artérielle et le cholestérol). Dernièrement, plusieurs bio-Marqueurs ont été examiné pour leur aptitude à améliorer la prédiction des maladies cardio-Vasculaires au-Delà des facteurs de risques traditionnels. L’intérêt de découvrir de nouveaux loci est incité notamment par les découvertes qui émergent des études d'association pangénomique (GWAS) qui permettent de tester l’association de variation génétique au risque de contracter une maladie commune. Les GWAS ont considérablement amélioré notre connaissance de l'architecture génétique des maladies cardio-Vasculaires, à ce jour plus de 50 variations génétiques sont formellement associées à des maladies cardio-Vasculaires, de même plus de 200 marqueurs génétiques seraient associés à des facteurs de risque cardiovasculaire traditionnels (p. ex. le taux sanguin des lipides, la pression artérielle, l’indice de masse corporelle et le diabète de type 2). Le succès remarquable de ces études d’association, qui a permis l’identification de nombreux bio-Marqueurs, a conduit à une réévaluation des données génétiques dans le but de définir des informations cliniquement utiles pour limiter et mieux prédire les risques de maladies, grâce à une application plus efficace des stratégies de prévention. Dans cette thèse, nous examinons tout d'abord une nouvelle approche pour étudier l'architecture génétique de l'hypertension artérielle (HTA; facteur de risque majeur des maladies cardiovasculaires prématurées), puis nous avons constitué plusieurs modèles pour prédire le risque de développer une maladie coronarienne (MC; type le plus commun de MCV), enfin nous avons déterminé une base génétique commune du principal prédicteur de complications cliniques des maladies coronariennes – l'athérosclérose subclinique - afin d'ajouter une valeur pronostique supplémentaire en plus des scores de risque traditionnels à différents âges. Nous avons estimé l'héritabilité de la première mesure de la pression artérielle systolique (PAS) à ~25%/~45% et à ~30%/~37% pour la pression artérielle diastolique (PAD) chez les sujets d’origine Européenne (N = 8901) et d’origine Africaine (N = 2860) faisant respectivement partie de la cohorte Atherosclerosis Risk in Communities (ARIC), en accord avec les études antérieures. Par ailleurs, nous avons développé un moyen de combiner un score de risque génétique (SRC) – somme des effets génétiques parmi un ensemble de marqueurs – avec une évaluation indépendante du risque clinique, en utilisant un système d'équations log-Linéaire. Nous avons employé cet outil à la prédiction de la maladie coronarienne (MC) dans la cohorte ARIC. L'ajout d'un score de risque génétique (SRG) à un score de risque clinique (SRC) améliore à la fois la discrimination et l'étalonnage des maladies coronariennes dans la cohorte ARIC, et révèle par la même comment cette information génétique influence l'évaluation des risques ainsi que l’approche clinique. Enfin, parmi 1561 cas et 5068 contrôles (de la présence ou non de calcifications coronaires), faisant partie de plusieurs ensembles de données cliniques et génétiques disponibles via la base de données NCBI de Génotypes et Phénotypes (dbGAP), nous avons constaté qu’une augmentation d'un écart-Type dans le score de risque génétique de 49 bio-Marqueurs de MC est associée à 28 % d’augmentation de risque de développer une athérosclérose coronarienne subclinique diagnostiquée à un stade avancé (p=1.43x10-16). Cette augmentation du risque est significative dans chaque catégorie d'âge (de 15 ans en 15 ans) (0,01 > p > 9.4x10-7) et a été remarquablement similaire dans toutes les catégories d'âge (test d'hétérogénéité p = 0.98). (...) / Complex diseases such as cardiovascular disease (CVD) are influenced by both genetic and environmental factors. Estimation of an individual’s cardiovascular risk usually involves measurement of risk factors correlated with risk of CVD (e.g. age, sex, smoking, blood pressure, and total cholesterol). Lately, several biomarkers have been evaluated for their ability to improve prediction of cardiovascular disease beyond traditional risk factors. The interest in novel loci is propelled notably by emerging discoveries from the advent of genome-Wide association studies (GWAS) of genetic variants associated with risk for common diseases. GWAS has greatly enhanced our knowledge of the genetic architecture of cardiovascular disease, yielding over 50 variants confirmed to be associated with CVD to date, as well as over 200 associated with traditional cardiovascular risk factors (e.g. lipids, blood pressure, body mass index, and type 2 diabetes mellitus). This recent and continuing success in discovering increasing numbers of robustly associated genetic markers has led to reassessment of whether genetic data can provide clinically useful information by refining risk prediction and moderating disease risk through a more efficient application of prevention strategies. In this thesis, we first address novel approach to survey the genetic architecture of hypertension (i.e. major risk factor for premature CVD), then construct risk prediction models for coronary artery disease (CAD; i.e. most common type of CVD) and finally establish a common genetic basis of the strongest predictor of clinical complications of CAD, subclinical atherosclerosis, to add incremental prognostic value above traditional risk scores across a range of ages. We show that, for first visit measurements, the heritability is ~25%/~45% and ~30%/~37% for systolic (SBP) and diastolic blood pressure (DBP) in European (N=8,901) and African (N=2,860) ancestry individuals from the Atherosclerosis Risk in Communities (ARIC) cohort, respectively, in accord with prior studies. Then we present a means to combine a polygenic risk score - genetic effects among an ensemble of markers - with an independent assessment of clinical risk using a log-Link function. We apply the method to the prediction of coronary heart disease (CHD) in the ARIC cohort. The addition of a genetic risk score (GRS) to a clinical risk score (CRS) improves both discrimination and calibration for CHD in ARIC and subsequently reveal how this genetic information influences risk assessment and thus potentially clinical management. Finally, Among 1561 cases and 5068 controls, from several clinical and genetic datasets available through the NCBI's database of Genotypes and Phenotypes (dbGAP), we found a one SD increase in the genetic risk score of 49 CAD SNPs was associated with a 28% increased risk of having advanced subclinical coronary atherosclerosis (p = 1.43 x 10-16). This increase in risk was significant in every 15-Year age stratum (.01 > p > 9.4 x 10-7) and was remarkably similar across all age strata (p test of heterogeneity = 0.98). We obtained near identical results and levels of significance when we restricted the genetic risk score to 32 SNPs not associated with traditional risk factors. Accordingly, common variation largely recapitulates the known heritability of blood pressure traits. The vast majority of this heritability varies by chromosome, depending on its length, and is largely concentrated in intronic and intergenic regions of the genome but widely distributed across the common allele frequency spectrum. Respectively, our proposed method to combine genetic information at established susceptibility loci with a nongenetic risk prediction tool facilitates the standardized incorporation of a GRS in risk assessment. (...)