Large-scale integrative studies of risk factors in coronary heart disease: from discovery to application

If medical studies can reliably demonstrate that a particular factor is relevant to heart disease, then this could have important implications for the prediction and prevention of disease, as, for example, is now the case with measurement and modification of blood LDL("bad")-cholesterol levels. Unfortunately, attempts using conventional approaches to identify risk factors that have cause-and-effect relationships with heart disease have often yielded costly failures in drug trials of new medicines. Similarly, only modest improvements have been achieved in the prediction of heart disease with incremental modifications to conventional approaches. Our research plan offers a fundamentally new approach to address these problems by combining the precision of molecular measurements with the power of large-scale population health studies. The deep integration of cutting-edge technologies for genetic and biochemical measurement technologies with extremely large and mature biomedical surveys ("epidemiological studies") of heart disease will yield studies that combine unprecedented power and detail. In particular, for 50,000 people with heart attacks and 50,000 controls, we have for each participant, already recorded extensive detail about: -genetic make-up (eg, up to one million genetic variants or "letters") -blood biochemistry (eg, up to a few hundred analytes) -lifestyle and other habits (eg, diet, physical activity, tobacco and alcohol consumption). In the most informative subsets of these participants, we will conduct state-of-the-art measurements to supplement this extensive existing information. A key advantage of blood measurement methods called "lipidomics" and "metabonomics" should be that, although they relate to biochemical processes likely to be relevant to the causation of heart attacks (ie, fat and sugar metabolism, respectively), they cast wide scientific nets. This should avoid premature assumptions about the identity of the precise factors that might emerge to be relevant to heart disease. Furthermore, as funding has already been awarded for "lipidomics" and "metabonomics" assays in 15,000 "control" participants in one of our studies (which we share with a study of new-onset type 2 diabetes), we will achieve major scientific synergy and cost savings by conducting concurrently the same assays in patients with new onset heart disease. To help harvest the complex and rich data that will emerge from our studies in a rigorous and principled manner, our team includes world leaders in biostatistics. We will build on innovative approaches that we have previously developed to help distinguish causal from non-causal factors in heart disease. The objective will be to discover entirely new causes of heart disease as well as to evaluate a variety of factors already suspected in heart disease, such as blood fats, sugar metabolism, "inflammation" (which is the body's response to injury), and nutritional factors. We will use the same databases (supplemented by additional sources of information) to develop and test a range of approaches that have potential to improve the prediction of first-onset heart disease, such as those that: -maximise accuracy of the prediction of heart disease, such as detailed scores containing genetic and biochemical information, which may be especially relevant to young people contemplating a lifetime of preventive therapy or to people with a strong family history of heart disease -promote efficiency for health services, eg, "sequential" screening, which initially involves comparatively simple tests for everyone, then focusing more costly and detailed measurements on individuals initially identified as being most likely to benefit from further assessment.