Alzheimer's diseaseBMJ 1994; 308 doi: https://doi.org/10.1136/bmj.308.6930.672 (Published 12 March 1994) Cite this as: BMJ 1994;308:672
- M Owen,
- M Liddell,
- P McGuffin
Alzheimer's disease accounts for around half of all cases of senile dementia. Its prevalence increases steeply with age, from less than 1% at age 65 to 15% in the ninth decade.1 Until last year advances in understanding the genetics of the disease had been confined to a rare variant known as familial Alzheimer's disease, which is of unusually early onset and segregates in an autosomal dominant fashion. Between 15% and 20% of familial Alzheimer's disease is caused by mutations in the amyloid precursor protein gene on chromosome 212 - a small fragment of this protein, known as ß amyloid is one of the main constituents of senile plaques. A second gene, which probably accounts for most familial Alzheimer's disease, has been mapped to the long arm of chromosome 14 but has not yet been identified.3
Familial Alzheimer's disease accounts for less than 1% of all cases of Alzheimer's disease. The genetics of the common variety are more complex. A large body of evidence indicates that this variety is familial, but unlike familial Alzheimer's disease it does not show a mendelian pattern of segregation, and less than half of all patients with the disease have a family history. Not surprisingly, the risk to relatives of patients depends on age: the cumulative risk of dementia in first degree relatives rises to 40% or more by the age of 85-90.4 Moreover, recent evidence of greater concordance in monozygotic than dizygotic twin pairs suggests that genetic factors contribute substantially to this increased risk.5,6 Attempts to identify the gene or genes responsible have been complicated by the fact that - as with other common disorders - the mode of transmission is obscure. Probably most cases are caused by the action of several different genes together with environmental factors, though the results of some analyses are consistent with a major gene with age dependent penetrance.7
Despite the difficulties recent work has shown a strong and apparently robust association of Alzheimer's disease with the (epsilon)4 allele of apolipoprotein E.*RF 8-11* The frequency of the (epsilon)4 allele is around 0.50 in affected individuals from families where several other members also have late onset disease. It is around 0.40 in patients unselected for the presence of a family history, including those in whom the diagnosis has been confirmed at necropsy. In control populations the frequency is around 0.12. The odds ratio is 6.4 for those carrying one or more (epsilon)4 allelles. The risk of Alzheimer's disease apparently increases and the mean age at onset decreases in direct relation to the number of (epsilon)4 alleles - showing that the (epsilon)4 allele acts in a partly additive fashion. Nevertheless, some 40% of people with Alzheimer's do not possess an (epsilon)4 allele so it is neither necessary nor sufficient to cause the disease and other aetiological factors and probably other genes must play a part. If we assume a polygenic and multifactorial mode of transmission12 then apolipoprotein (epsilon)4 accounts for about 17% of the variance in the liability to develop the disorder in the general population.
A more practical issue, with ethical implications, is whether apolipoprotein E genotyping can be used for predictive or diagnostic purposes. In multiply affected families more than 90% of members who are homozygotes for (epsilon)4 seem to develop the disorder by the age of 80.10 Calculations of the true risk in the general population are, however, complicated by the fact that too few octogenarians have yet been studied. The only published study of a consecutive series of patients reported a risk to 80-90 year olds with the (epsilon)4/(epsilon)4 genotype of 84%, but that series included only 37 patients aged over 80. Moreover, it is not possible to conclude from this case-control study (as some have done13) that the risk to (epsilon)4/(epsilon)4 homozygotes in the general population of 80-90 year olds is as high as 84%. If we adopt a bayesian approach and assume a population prevalence of Alzheimer's disease of 15% in 85 year olds and a frequency of (epsilon)4/(epsilon)4 homozygotes of 0.03 in the general population and 0.13 in Alzheimer's disease, then we can calculate that the probability of someone with the (epsilon)4/(epsilon)4 genotype having Alzheimer's disease at age 85 is 43%.11
The predictive value of testing for (epsilon)4 would be even lower when younger populations were tested - because fewer people will reach the age of risk. For example, if we assume that 35 year olds have a 2% chance of developing Alzheimer's disease in their lifetime then only 6.4% of (epsilon)4/(epsilon)4 homozygotes of that age will develop the disease.14 These calculations are further complicated by the fact that fewer (epsilon)4/(epsilon)4 homozygotes are likely to survive long enough to develop Alzheimer's disease because of their risk of cardiovascular disease.
The allele may also be a risk factor for multi-infarct dementia - which is not surprising, given that those with this allele are at increased risk of atherosclerosis. This second association might increase the predictive value of apolipoprotein E genotyping as a test for dementia, but on current data the risk of dementia at age 85 would still be only about 65%, assuming that being an (epsilon)4 homozygote confers a similar level of risk of multi-infarct dementia. Testing is unlikely to be of value in the differential diagnosis of senile dementia since multi-infarct dementia and Alzheimer's disease between them account for over 90% of cases.
The really important issue is how apolipoprotein E contributes to the pathogenesis of Alzheimer's disease. Possibly (epsilon)4 acts as a neutral marker for another gene with which it is in linkage disequilibrium and which predisposes to Alzheimer's disease; but, the presence of apolipoprotein E in both senile plaques and neurofibrillary tangles, together with a possible contribution to the response to neuronal injury, points to it playing a direct part in pathogenesis.13 Possibly apolipoprotein E4 might increase the rate at which deposits of amyloid accumulate; or it might reduce the stability of microtubules, leading to the accumulation of neurofibrillary tangles. Whatever the mechanism, its elucidation will ultimately offer a real hope for effective treatment.
Our research in Alzheimer's disease is supported by a grant from the Mental Health Foundation.
Log in using your username and password
Log in through your institution
Register for a free trial to thebmj.com to receive unlimited access to all content on thebmj.com for 14 days.
Sign up for a free trial