How Could Use of Genetic Markers Prevent Coronary Heart Disease Events?

Samuli Ripatti; Veikko Salomaa

Disclosures

Personalized Medicine. 2013;10(8):769-771. 

In This Article

Search for Coronary Heart Disease Biomarkers

Coronary heart disease (CHD) is the number one cause of death in industrialized countries.[1] This makes prevention of CHD a public health priority. The key for successful prevention is early identification of high-risk individuals and the possibility of successful interventions. CHD is a promising target as interventions are possible through lifestyle changes in dietary, exercise and smoking habits, and lipid-lowering and antihypertensive medications for high-risk individuals.

However, identification of high-risk individuals using traditional risk algorithms, such as the Framingham risk score, is far from perfect. In fact, more than half of heart disease events occur in individuals with estimated risk at low or average levels. As an example, in the Finnish FINRISK cohorts, 69% of the new incident cases occurred in individuals with 10-year risk below 20%, which is a typically used clinical threshold for intervention.

This has resulted in the search for novel biomarkers to potentially help in risk estimation and development of statistical concepts to evaluate their potential. A traditional metric to evaluate a new biomarker is its potential to discriminate cases from noncases in a population sample. Area under the receiver operating characteristic curve and the corresponding c-statistic provide a summary score for the balance between the sensitivity of the biomarker to detect a case and 1-specificity. While a c-statistic of 0.5 corresponds to flipping a coin to give a risk prediction, 1.0 corresponds to perfect discrimination and risk estimation. Another approach is to divide the individuals into risk categories based on their predicted risk with and without the new biomarker, and compare the proportion of cases reclassified into a higher risk group and controls reclassified into a lower risk group with those who are reclassified in the 'wrong' direction. Net reclassification improvement (NRI) compares these proportions and calculates the percentage of individuals reclassified in the 'right' minus the 'wrong' direction.

Although the number of tested biomarkers is large and includes many circulating biomarkers and imaging summary measures, few have shown a gain in discrimination and/or reclassification. The most promising markers include coronary calcium score and intima–media thickness, which have shown improvements in c-statistics around 1–5% and NRIs 8–25%.[2–6]

processing....