Diabetes biomarkers

Published on June 24, 2013   32 min

A selection of talks on Clinical Practice

Please wait while the transcript is being prepared...
Hello, my name is Naveed Sattar. I am professor of metabolic medicine at the University of Glasglow, and I am going to give you an overview of diabetes biomarkers.
So biomarkers, what are they? And what are they for? Biomarkers are really any measure, whether it's biological material or even a clinical characteristic which gives you insight into either disease, pathogenesis, prediction of complications or progression of follow up. Or it can actually give you an insight into the response of an individual to a particular therapy or intervention. So biomarkers can range from simple things such as age, social class, but of course, most people understand the term biomarkers in terms of some blood measure or some biochemical parameter, for example, cholesterol or glucose.
If we take cardiovascular disease as an example of how they have used biomarkers, well, we are now at a point where we know that age, blood pressure, smoking, gender, lipids, and the presence of diabetes gives insight into cardiovascular risk. And from this, we have developed cardiovascular risk scores, ranging from the original Framingham risk scores through to the risk score in Europe, as well as more sophisticated risk scores which have added other potential parameters which we will discuss in the next couple of slides. There is big-scale epidemiology on these routine biomarkers in cardiovascular disease in terms of prediction. And some of the lessons we have learned from cardiovascular disease in terms of predicting disease are relevant to diabetes research.
If we take first the relationship of lipids to cardiovascular outcomes, this has now been established in huge data sets, culminating in collaboration of multiple cohorts called Emerging Risk Factor Collaboration, which reported the lipid data in 2009 in JAMA, and shows clearly that non-HDL or LDL is strongly and linearly related to hazard ratio cardiovascular events. HDL is inversely related, and triglyceride is positively related in analysis adjusted for age and sex only, whilst LDL or non-HDL and HDL remain associated with CHD in the same pattern one to further adjusted for non-lipid and lipid factors. Triglyceride actually shows no association once other lipid parameters and another risk factors are accounted for. This suggests that actually, LDL and HDL or total cholesterol and HDL are all that we really need in terms of predicting cardiovascular disease, and that triglyceride does not add prediction over and above other lipid markers. And in actual fact, further evidence published in other papers which suggests that triglyceride is in fact a stronger risk factor in the development of diabetes than it is for cardiovascular disease.