Genetic epidemiological studies of coronary heart disease

The recent publication of the human genome sequence is widely thought to offer the opportunity for a radical change in our understanding of a variety of common human diseases. One particular hope is that the new information available from the genome sequencing effort will facilitate the conduct of population genetic studies, which will discover the genetic variants responsible for ‘complex’ or ‘polygenic’ (i.e. resulting from the action of more than one gene) diseases. In this review I attempt to put this aspiration into perspective for coronary heart disease (CHD). Firstly, I consider what is known thus far regarding the ‘genetic architecture’ of CHD. Secondly, I discuss the implications of this ‘genetic architecture‘, and of certain population genetic issues, for study design. Thirdly, I consider the reasons why the results of genetic-epidemiological studies of CHD to date have tended to be discrepant, and explore strategies for increasing the reliability of such studies in the future. Finally, I consider how genetic-epidemiological studies of adequate size may be uniquely well placed to resolve particularly important and difficult controversies regarding the causative nature of hypothesized novel risk factors for CHD. There are a number of reasons why genetic studies of complex diseases such as CHD have moved to a central position in the thinking of both geneticists and epidemiologists. Firstly, genetic studies offer the opportunity to identify determinants of disease that are very likely to be causative. This is because genotypes are unchanged throughout life, and therefore not susceptible to confounding via mechanisms related either to the presence of disease itself or the body’s response to disease (in the way that measurement of hypothesized risk factors in plasma, for example, could be). Secondly, the current pharmacological armamentarium for complex diseases consists of drugs which act on a very small number (several hundred) of the 30 000 or so genes in the human genome. Identification of novel genes that contribute to CHD risk could therefore lead to new therapeutic targets with strong molecular underpinning. Thirdly, identification of risk genotypes might enable more accurate determination of an individual’s risk of CHD than is currently possible from measurement of known risk factors. This is partly because genotypes could measure the activity of novel or otherwise unmeasurable biological pathways, and partly because genotypes would not be susceptible to short-term fluctuations and measurement error, in contrast to measurements of plasma and other quantitative phenotypic risk factors.