Purpose: Genetic variation of a common genetic polymorphism at the structural locus for apolipoprotein E (apoE) has been associated with risk of elevated serum lipids, coronary artery disease, and Alzheimer disease, all of which are multifactoral disorders and are inherited in a complex fashion. Although the apoE polymorphism is only one risk factor in a complex pattern of inheritance, it gives us a foothold into the understanding of the genetic architecture of these disorders. However, apoE also exemplifies the complications that can arise in the use of genetic markers to predict disease. This paper considers the study of apolipoprotein E genotype effects as an example of the following analytical complications: changes in allele frequencies and allele effects with age, pleiotrophic effects of genetic loci, the existence of more than one alternative allele at a locus, and the expectation of a variety of interactions.
Methods: Publications that exemplify these complications are cited and discussed. No original analyses are presented.
Results: There is evidence that the relative frequency of the apoE epsilon4 allele declines with age after the sixth decade of life and that the effects of the APOE allele on lipids also may be age-dependent. Grouping of genotypes may not accurately characterize the effects of individual genotypes. Interactions between APOE genotype and a number of factors, including family history of dementia, are demonstrated for the effects of APOE in Alzheimer disease as well as for the effects of apoE on serum lipids.
Conclusion: Careful attention must be paid to these and other analytical issues when genotype is a predictor.