SPECIFIC AIMS AND HYPOTHESES
The National Heart, Lung, and Blood Institute (NHLBI) initiated in 1992 a large, population-based multicenter family study of the genetic and non-genetic determinants of coronary heart disease (CHD), atherosclerosis, and cardiovascular disease (CVD) risk factors, the Family Heart Study or FHS, (RFP NHLBI-HC-91-07-09). During the initial funding period (May 1992 - April 1996), extensive questionaire information on family and medical histories were collected on large numbers of random and coronary prone families (Phase I). Based upon this information, in Phase II, the highest CHD risk families along with a random sample of families were extensively clinically examined for CHD risk factors, including biochemical profiles, lipid profiles, ECG, pulmonary function, ultrasonography (intima-media wall thickness, IMT, of the carotid artery by B-mode ultrasound), as well as life style information. DNA was extracted and stored for later molecular studies, as were blood specimens for future risk factor studies.
This multicenter study has now begun to utilize the rich resource of family data collected, by undertaking molecular genetic, and genetic epidemiology studies. This research seeks to identify and characterize genes, lifestyle and behavioral characteristics, and advance our understanding of how they interact with one another in the genesis of clinical and pre-clinical disease. CHD, atherosclerosis, and CVD risk factors all have a significant genetic component, but share a lack of phenotypic specificity, multigene determination, and etiological heterogeneity. These features contribute to a complex picture eluding conventional approaches of classical Mendelian genetics. Consensus has emerged, however, that there is a reasonable likelihood that such genetic determinants can be identified by a combination of molecular marker technology and robust, model-independent methods to infer genetic linkage and association.
The general plan is to perform linkage analysis on large numbers of sibpairs ascertained for CHD or for excess atherosclerosis, first using candidate genes, and later extending the search to the entire genome. Once linkage is found to known genes, direct searches for molecular variants will be followed by tests of association in case-control comparisons. The strategy will closely parallel that used successfully by the molecular genetics lab to implicate angiotensinogen (AGT) in human hypertension. The ultimate goal of the work is the development and dissemination of new information on the familial determinants of CHD and atherosclerosis, and their relative and attributable risks. We believe that the already collected FHS data set, which is both extensive and of high quality, together with carefully banked DNA, provides a unique conjuncture to identify several major genetic determinants of CHD and atherosclerosis, and their modulation by other genetic and environmental influences. Statistical power also exists to evaluate other genetic (and environmental) determinants of smaller magnitude of effect. Further, this can be accomplished optimally in the organizational environment of a single, standardized population-based family study.
The premise underlying the FHS Study is that molecular variation at some loci predisposes individuals to the development of Coronary Heart Disease (CHD), Atherosclerosis, and a number of important intermediate phenotypes and novel risk factors. Multifactorial determination and heterogeneity suggest that several genes may account for a predominant fraction of this genetic component. The primary objective of this phase of the research is to identify such genetic determinants, the non-genetic determinants, and to investigate how these genes and environmental factors act together to produce clinical and pre-clinical disease. This strategy takes advantage of a highly informative and already collected population-based Family Heart Study data set on 583 random sample families and 649 coronary-prone families, including carefully banked DNA. The plan includes: state-of-the-art molecular genetic studies including genome-wide searches, imaginative genetic epidemiological studies, with attention to the interplay between these two approaches, and carry out gene characterization studies and gene-gene and gene-environment interaction studies in defined populations.
A.1 Specific Aims
A.1(a) Test for linkage of 30 candidate genes and 240 anonymous markers to CHD, using an affected sibpair approach with 124 affected sibpairs in 88 sibships, where CHD has been validated in all affected sibs. We hypothesize that there are genes that influence CHD and risk factors that are currently not identified as candidate genes because the proteins have not been discovered yet, or are not thought to be related to CHD. It is the objective of the genomic screens using anonymous markers to help localize such genes.
A.1(b) Test for linkage of 30 candidate genes and 240 anonymous markers to CHD-related traits (including atherosclerosis and several intermediate phenotypes), using a Quantitative Trait Locus (QTL) approach and a large set of 2,492 sibpairs in 551 sibships, each sibship with at least two full sibs examined in Phase II of FHS, where at least one of them is above the 85th percentile of the age-sex-specific distribution of an Individual Risk Score (IRS) - an index of risk for CHD. A two-stage linkage design is proposed, where a "coarse" map is used with half the sample to detect linkages in stage-1, and a "dense" map is used with the other half sample in stage-2 for replication and finer mapping.
A.1(c) Characterize these CHD-related genes by genotyping a random sample of 200 subjects, all of whom already have extensive biochemical tests performed, and 72 non-CHD sibs of the CHD gene carriers. Using the genes identified by this study or others, ascertain: (1) the frequency of the diseasepromoting gene in the general population; (2) the proportion of individuals exhibiting the adverse phenotype that carry this gene; and, (3) the proportion of gene carriers who express the adverse phenotype at a given age. Determine the precise role of the gene variant in the promotion of the adverse phenotype, and draw inferences about pathophysiology. Determine to what degree expression or penetrance are influenced by the effects of other genes, age & sex, or environmental factors. Determine the biochemical and physiological differences of gene carriers compared to non carriers. Among those who do carry this gene variant, ascertain the differences between individuals who express the adverse phenotype versus those who do not.
A.1(d) Investigate the Genetic Epidemiology of Atherosclerosis, its Sequelae, and Intermediate Phenotypes, with emphasis on evaluation of hypotheses in the random and nonrandom (high risk) samples of (mostly) Caucasian families:
(1) Characterize the familial aggregation, temporal trends in familial effects, and major gene effects on CHD, carotid atherosclerosis as measured by IMT, peripheral arterial disease, and arterial stiffness.
(2) Evaluate familial clustering of hypertension, dyslipidemias, and metabolic CVD risk factors, with emphasis on the detection of pleiotropic effects using multivariate methods.
(3) Evaluate and describe the familial aggregation of psycho-social risk factors of CHD (measures of hostility, social support, and job stress), and evaluate especially their role as covariates in the promotion or inhibition of CHD gene effects.
(4) Evaluate the familial aggregation, temporal trends, and major gene effects on asthma and on measures of pulmonary function (FVC, FEV1).
(5) Test for linkage of 30 candidate genes and 240 anonymous markers to intermediate phenotypes, and traditional and novel risk factors (e.g., hemostasis, lipoprotein[a]) of cardiovascular disease, and informative familial clusters thereof.
A.1(e) Explore the Genetic Epidemiology of atherosclerosis and its intermediate phenotypes in African Americans, with emphasis on generation of hypotheses. The considerably smaller numbers of families and of family sizes precludes a definitive evaluation of hypotheses in the African American family data. Nonetheless, these data are valuable for the exploration of basic familial patterns in African Americans. Accordingly, we propose to:
(a) Investigate familial aggregation of carotid atherosclerosis, and their risk factors of greatest relevance to African Americans.
(b) Contrast the familial patterns in the Caucasians and African Americans. Potentially important findings will be re-evaluated using the combined sample of Caucasians and African Americans.
A.1(f) Test for interactions between the genes found and a large number of non-genetic determinants of CHD and atherosclerosis. Assess the interplay of genes and environment in the development of CHD, atherosclerosis, and a variety of intermediate phenotypes.
A.1(g) Assess Population/Regional differences in the Genetic Epidemiology of CHD, atherosclerosis, intermediate phenotypes and novel risk factors. The FHS data set offers an opportunity to evaluate possible differences among the populations inducted at different field centers, as well as to characterize any observed differences (under the hypothesis that there are no genetic differences among the Field Centers).
We believe that the proposed study design represents one of the most powerful designs for achieving these goals, and that the data already collected at considerable cost provides a unique opportunity to identify several major genetic determinants in an area of extraordinary public health importance. The cost to society of not performing the proposed research is potentially greater than the dollar cost of the proposed research. Conversely, the potential benefits to society of performing this research are significant. We are aware that, despite the favorable circumstances, there is a non-zero probability that this study will fail to identify any new genes unambiguously. If such an unexpected outcome should take place it will have far-reaching implications for designing future studies of this type.
A.2 Overview of Hypotheses, their Priority, and Time Line.
The following table provides a summary of the study questions to be addressed in Phase III. Each of the aims listed above is broken down into more specific study questions (or "hypotheses"). To aid in the review, the study questions are presented grouped into hypotheses that can be addressed on extant FHS data, and the corresponding questions that will be addressed as the molecular genetics information has become available over the course of the first two years of this study. An almost overwhelming number of hypotheses could be generated from the large FHS data base. The list that follows is extensive but realistic as the work scope for this study, in that it was selected based on scientific merit, and by establishing priorities for the work to be conducted over the course of four years. This ranking indicates the priority each it will receive in the allocation of time and resources in the context of the work proposed here. For each study question the proposed analytic method is provided in outline form, as well as its place in the study time line and the group of investigators responsible for its completion. The time line, shown here only in one year increments, reflects both the availability of data and the scientific importance of the topic.
The assignments to the investigators shown in this table are a result of careful planning in the process of preparing this application, and represent the input as well as the explicit commitments from the collaborating scientists. Accordingly, the assignments to various analytic and publication projects reflect a balance of scientific priority, the complexity of the task and type of expertise required, a full utilization of the study's resources, and a mechanism to avoid duplicative efforts. Each project is uniquely identified with a lead investigator responsible for the project's quality and timely progress, and monitored by the Publications Committee.