Multiple biomarkers refine cardiovascular disease risk

1 June 2008

A biomarker is any form of biological molecule (present in tissues, blood or other body fluids) that is associated with a particular pathological or physiological state, and which can be used to indicate or measure the presence of a substance, event or condition. For example, detection and measurement of a specific protein might indicate the presence and /or progress of a certain disease.

For cardiovascular disease, it is already possible to assess an individual’s risk based on a number of clinical characteristics and biomarkers such as levels of blood cholesterol, all of which have been reliably shown to affect risk. More recently, additional biomarkers have been reported to correlate with increased cardiovascular risk. New research published in the New England Journal of Medicine has reported that rthe addition of multiple biomarkers can improve cardiovascular disease risk prediction. Accurate risk stratification is important in public health because it allows clinicians to identify individuals at greatest risk of the disease, who may then be targeted for preventative interventions.

The researchers looked at data gathered from a cohort of over 1100 elderly Swedish men (including 661 without detectable cardiovascular disease at the start of the study) and their ten-year outcomes [Zethelius B et al. (2008)N Engl J Med 358(20):2107-16]. Over this period, a total of 315 subjects died, 136 of them from cardiovascular disease. Four biomarkers were considered: 

  • troponin I (a biomarker for myocardial heart cell damage)
  • N-terminal pro–brain natriuretic peptide (a biomarker for left ventricular heart dysfunction)
  • cystatin C (a biomarker for renal / kidney failure )
  • C-reactive protein (a biomarker for inflammation)

These biomarkers reflect pathological processes that have been linked with an increased risk of cardiovascular disease and death, but are not used in current risk assessments. The question was whether or not the use of biomarker data could improve risk stratification beyond that achieved by a standard assessment using established risk factors for cardiovascular disease: age, systolic blood pressure, use of anti-hypertensive or lipid-lowering treatments (for example, beta-blockers or statins), total and high-density lipoprotein blood cholesterol levels, presence or absence of diabetes, smoking status, and body-mass index.

Statistical analysis suggested that all the selected biomarkers showed a significant degree of predictive ability with respect to the risk of death from cardiovascular disease, and especially when combined in a predictive model with established risk factors. This was true both for the sample as a whole, and for the sample sub-group in whom there was no detectable cardiovascular disease at the start of the study.

Previous studies have failed to demonstrate a statistical improvement in risk stratification from the inclusion of biomarker levels in the risk assessment criteria. The authors propose that their findings may have been influenced by the increased age of the study participants compared with those in earlier studies; the risk associated with the established risk factors decreases with increasing age, so that the information provided by the new biomarkers becomes more significant. They also note that previous work had evaluated biomarkers singly for their effect on risk prediction, whereas they showed that none of the biomarkers had a significant effect on risk prediction when individually added to the established risk criteria.

Comment: In demonstrating that the use of combined data from a number of different biomarkers can successfully refine risk stratification, despite the fact that each biomarker taken singly did not do so, this study may pave the way for further developments in the identification of useful combinations of biomarkers associated with disease risk, for risk prediction. Further research will be necessary to determine whether or not these particular biomarkers are robust enough to be used in standard clinical practice for the prediction of death from cardiovascular causes, based on their performance in different cohorts (women, younger people, and different ethnic groups). However, it may be that risk prediction of many different diseases could potentially be improved by the inclusion of multiple new biomarkers in risk assessment, if these biomarkers can be identified and validated.

More from us

Genomics and policy news