Usefulness of genetic polymorphisms and conventional risk factors to predict coronary heart disease in patients with familial hypercholesterolemia

Abstract

Familial hypercholesterolemia (FH) is an autosomal dominant disorder with an associated high risk of coronary heart disease (CHD). The considerable variation in age of onset of CHD in patients with FH is believed to arise from conventional risk factors, as well as genetic variation other than in the low-density lipoprotein receptor gene. The degree to which currently known genetic variants can improve the prediction of CHD risk beyond conventional risk factors in this disorder was investigated. Fourteen genetic variants recently identified for association with CHD in a Dutch FH population were considered. Prediction models were constructed using Cox proportional hazards models, and predictive value was assessed using a concordance statistic (c statistic). A total of 1,337 patients with FH were completely genotyped for all genetic variants. Hazard ratios of the genetic variants ranged from 0.61 to 0.74 and 1.24 to 2.33. The c statistic of the CHD prediction model based on genetic variants was 0.59, denoting little discrimination. The model based on conventional risk factors had a c statistic of 0.75, denoting moderate discrimination. Adding genetic test results to this model increased the c statistic to 0.76. In conclusion, the contribution of 14 genetic variants to the prediction of CHD risk in patients with FH was limited. To improve genome-based prediction of CHD, larger numbers of genetic variants need to be identified that either on their own or in gene-gene interaction have substantial effects on CHD risk.