Rationale and design of the LURIC study - a resource for functional genomics, pharmacogenomics and long-term prognosis of cardiovascular disease

Abstract

Background and aim: Coronary artery disease (CAD), arterial hypertension and Type 2 diabetes mellitus are common polygenetic disorders which have a major impact on public health. Disease prevalence and progression to cardiovascular complications, such as myocardial infarction (MI), stroke or heart failure, are the product of environment and gene interaction. The LUdwigshafen RIsk and Cardiovascular Health (LURIC) study aims to provide a well-defined resource for the study of environmental and genetic risk factors, and their interactions, and the study of functional relationships between gene variation and biochemical phenotype (functional genomics) or response to medication (pharmacogenomics). Long-term follow-up on clinical events will allow us to study the prognostic importance of common genetic variants (polymorphisms) and plasma biomarkers. Setting: Cardiology unit in tertiary care medical centre in south-west Germany. Study design: Prospective cohort study of individuals with and without cardiovascular disease at baseline. Patients and methods: LURIC is an ongoing prospective study of currently > 3300 individuals in whom the cardiovascular and metabolic phenotypes CAD, MI, dyslipidaemia, hypertension, metabolic syndrome and diabetes mellitus have been defined or ruled out using standardised methodologies in all study participants. Inclusion criteria for LURIC were: German ancestry (limitation of genetic heterogeneity), clinical stability (except for acute coronary syndromes [ACSs]) and availability of a coronary angiogram (this inclusion criterium was waived for family members provided that they met all other inclusion and exclusion criteria) Exclusion criteria were: any acute illness other than ACSs, any chronic disease where non-cardiac disease predominated & a history of malignancy within the past five years Exclusion criteria were pre-specified in order to minimise the impact of concomitant non-cardiovascular disease on intermediate biochemical phenotypes or on clinical prognosis (limitation of clinical heterogeneity). A standardised personal and family history questionnaire and an extensive laboratory work-up (including glucose tolerance testing in non-diabetics and objective assessment of smoking exposure by determination of cotinine plasma levels) was obtained from all individuals after informed consent. A total of 115 ml of fasting venous blood was sampled for the determination of a pre-specified wide range of intermediate biochemical phenotypes in serum, plasma or whole blood, for leukocyte DNA extraction and immortalisation of B-lymphocytes. Biochemical phenotypes measured included markers of endothelial dysfunction, inflammation, oxidative status, coagulation, lipid metabolism and flow cytometric surface receptor expression of lympho-, mono- and thrombocytes. In addition, multiple aliquots of blood samples were stored for future analyses. Results: A total of 3500 LURIC baseline measurements were performed in 3316 individuals between July 1997 and January 2000. The baseline examination was repeated within a median of 35 days in 5% of study participants (n = 166, including a third examination in 18 after a median of 69 days) for pharmacogenomic assessment of lipid-lowering therapy and for quality control purposes. A five-year follow-up on major clinical events (death, any cardiovascular event including MI, stroke and revascularisation, malignancy and any hospitalisation) is ongoing. The clinical phenotypes prevalent at baseline in the cohort of 2309 men (70%) with a mean age of 62 ± 11 years and 1007 women (30%), mean age 65 ± 10 years, were angiographically-documented CAD in 2567 (79%), MI in 1368 (41%), dyslipidaemia in 2050 (62%) with hypercholesterolaemia = 240 mg/dl (27%), hypertriglyceridaemia = 150 mg/dl (44%) and HDL-cholesterol = 35 mg/dl (38%) in individuals not treated with lipid-lowering agents, systemic hypertension in 1921 (58%), metabolic syndrome in 1591 (48%), Type 2 diabetes in 1063 (32%) and obesity defined by body mass index = 30 kg/m2 in 770 (23%). Control patients in whom CAD had been ruled out angiographically were five years younger than those with CAD (59 ± 12 and 64 ± 10 years, respectively; p < 0.001), twice as often females (48% compared to 25% females in the CAD group, p < 0.001) and had significantly less cardiovascular risk factors than individuals with CAD. The prevalence of specific cardiovascular risk subsets in LURIC, such as the elderly (= 75 years), was 375 (11%), while 213 (6%) were young adults (< 45 years) and 904 (27%) were postmenopausal women (90% of all females). A low risk status (= 1 out of the four traditional risk factors: dyslipidaemia, smoking, hypertension and diabetes mellitus) was identified in 314 (9%) individuals of the entire cohort (5% in CAD and 26% in controls, p < 0.001) and 97 (3%) carried none of the four risk factors (1% in CAD and 9% in controls, p < 0.001).A total of 35 aliquots (serum, plasma and red blood cells) were kept for long-term storage and future analyses at -80°C. Multiple stocks of whole blood and extracted DNA are available for ongoing and future genomic studies, including aliquots of B-cells frozen in liquid nitrogen for transformation into lymphoblastoid cell lines. All phenotypic data are stored in a database currently carrying 1710 variables on each individual. Conclusions: LURIC is a study to contribute to the identification and assessment of environmental and genetic factors for atherosclerosis and related metabolic diseases. Major efforts have been undertaken to establish reliable phenotypes for the cardiovascular and metabolic diseases CAD/MI, hypertension and diabetes mellitus. Phenotypic characterisation included a standardised clinical and an extensive laboratory work-up. DNA and plasma aliquots were stored for future use. Furthermore, immortalised cell lines offer a renewable source of DNA. The LURIC database and tissue bank is a resource to elucidate functional relationships between genetic variation and biochemical phenotype (functional genomics), to study genetically modified responses to medication (pharmacogenomics) and to study the prognostic importance of biomarkers of cardiovascular and metabolic disease.