R. Duara et al., ALZHEIMERS-DISEASE - INTERACTION OF APOLIPOPROTEIN-E GENOTYPE, FAMILYHISTORY OF DEMENTIA, GENDER, EDUCATION, ETHNICITY, AND AGE-OF-ONSET, Neurology, 46(6), 1996, pp. 1575-1579
We evaluated 197 patients with predominantly late-onset Alzheimer's di
sease (AD) who belonged to several ethnic groups and analyzed the rela
tionship of age of onset of AD to the presence or absence of several r
isk factors in this entire group of patients. The apolipoprotein E (ap
oE) epsilon 4 allele frequency, which was 29% in all patients (compare
d with the reported population mean of 13.7%, p < 0.001, did not vary
significantly between ethnic groups but declined significantly with in
creasing age. The apoE epsilon 2 allele frequency was 3%, compared wit
h the reported population mean of 7.4% (p = 0.001). The frequency of a
positive family history of dementia in first-degree relatives (FH+) (
overall 45%) did not vary significantly between ethnic groups. ApoE ep
silon 4-positive (epsilon 4+) patients tended to have a higher FH+ rat
e (58%) than apoE epsilon 4-negative (epsilon 4-) patients (40%) (p =
0.02). When the potential risk factors of gender, education, FH+ statu
s, and epsilon 4+ status were examined together in a multiple linear-r
egression analysis, FH+ and epsilon 4+ status (but not gender or educa
tion) were significant (they were both associated with an earlier age
of onset of AD). In a post-hoc analysis, we found a reduced age of ons
et in women, but not men, who were both FH+ and epsilon 4+. Additional
ly, those probands who were epsilon 4+ were more likely to inherit the
disease from their mothers than their fathers. The mechanism by which
epsilon 4+ and FH+ status operate as risk factors may be by their eff
ect on the age of onset of AD.