Genotypes and phenotypes for apolipoprotein E and Alzheimer disease in theHonolulu-Asia Aging Study

Background: The utility of apolipoprotein E (ApoE) type as an indicator of genetic susceptibility to Alzheimer disease (AD) depends on the reliability of typing. Although ApoE protein isoform phenotyping is generally assumed equivalent to genotyping from DNA, phenotype-genotype differences have been reported, Methods: ApoE genotype and phenotype results were examined for 3564 older ( ages 71-93 years) Japanese-American male participants of the Honolulu-Asia Aging Study, an ongoing population-based study of aging and dementia. Results: Both methods demonstrated similar associations of ApoE type with A D: a direct association with ApoE4 and a less dramatic inverse association ApoE2. Advanced age did not appear to influence the ApoE4-AD association. T he association with AD among ApoE4 homozygotes [odds ratio (OR) = 14.7] was higher than expected based on an observed OR of 2.0 in heterozygotes. Phen otype-genotype nonconcordance was more frequent for ApoE2 than for ApoE4. T he ApoE2 phenotype occurred at a frequency of 7.9% vs a genotype frequency of 4.9%, corresponding to a probability of 56% that an individual with ApoE 2 phenotype had the same genotype. Conclusions: Whereas E4 and E2 phenotypes and genotypes were comparably ass ociated with AD, neither method would be expected to substantially improve the efficiency of case finding in the context of population screening beyon d prediction based on age and education. Nonconcordance of phenotype and ge notype was substantial for E2 and modest for E4 in this population. The Apo E4-AD association was independent of age; (C) 2000 American Association for Clinical Chemistry.