Genetic Markers for Alzheimer Disease

Are They Ready for Prime Time?

Gary W. Small, MD


January 31, 2011

Genome-Wide Association Study of CSF Biomarkers Abeta1-42, t-tau, and p-tau181p in the ADNI Cohort

Kim S, Swaminathan S, Shen L, et al
Neurology. 2011;76:69-79


In recent years, cerebral spinal fluid (CSF) levels of beta amyloid and tau have been used as early diagnostic indicators for the development of Alzheimer disease in the future.[1] Although the impact of candidate genes on such markers has been studied, information about the results of the genome-wide association study (GWAS) is limited.

In this investigation, the effect of single nucleotide polymorphisms (SNPs) were assessed under an additive genetic model on each of 5 CSF biomarkers. After first correcting P values of each SNP for multiple comparisons, the researchers looked at SNPs with corrected P values < .01 to identify candidate SNPs that are associated with CSF biomarkers, and then looked at SNPs with uncorrected P values < 10-5 to identify any additional potential candidates.

The results indicated that 4 SNPs in the regions of the APOE, LOC100129500, TOMM40, and EPC2 genes reached genome-wide significance for associations with 1 or more of the CSF biomarkers, and several others were identified as potential candidates.


Finding useful genetic markers for age-related diseases like Alzheimer disease is a challenge. Given the risk for false-positive results, scientists are careful to guard against spurious findings. This recent GWAS report is appropriately cautious in its conclusion about the need for replication of the results, and the investigators set conservative levels for establishing the statistical significance of any finding.

Of note, the investigators used data from the large, multisite Alzheimer Disease Neuroimaging Initiative, which is focusing primarily on neuroimaging measures that might improve diagnosis and early detection of Alzheimer disease. The collaborative study also collects CSF biomarkers on a subset of subjects, and these biomarkers have been found to predict whether patients who do not have dementia will have progressive cognitive decline.[2] A useful genetic marker using GWAS of CSF biomarkers is a reasonable strategy because presumably most people would prefer the less invasive procedures available to determine their genetic profile over a spinal tap, which can cause headache or infection.

Although the findings in this study are of considerable interest, their application to everyday clinical practice is limited. For individuals seeking genetic information about Alzheimer disease, the predictive value is unclear, unless the individual is from a family carrying one of the rare autodominant APP or presenilin mutations.[3]

For example, for many years, we have known that information about APOE genotype is neither necessary nor sufficient to predict a diagnosis of Alzheimer disease. Some people with the APOE4 mutation, which has been associated with an increased risk for disease, will never get the disease in their lifetime, and many without the APOE4 mutation will get the disease. That is why experts have been reluctant to recommend these types of "risk" tests as individual predictive tools -- there is concern that some people would be falsely alarmed by their test results while others would be falsely reassured. (It is interesting to note, however, that systematic studies of people who choose to obtain their APOE genetic results do not seem to show any obvious negative psychological consequences from these results.[4])

Even though many people are interested in their genetic risk profiles, and several companies are now providing such information to consumers, the value of predictive testing -- whether through the use of genetic findings, neuroimaging measures, or other biomarkers -- will only accelerate when such predictive measures are tied to prevention treatments. In my experience, most people are not interested in finding out years in advance whether they will get Alzheimer disease unless there is some treatment attached to that knowledge. Those for whom knowledge may be a motivation to adopt healthier brain-protective lifestyles are exceptions.

However, were a genetic test linked to a prevention treatment, Alzheimer disease might be managed in a way similar to other age-related illnesses that currently have available surrogate markers for the disease. For example, if a treatment that could delay the onset of dementia were tied to a predictive genetic test, people with certain risk factors such as age or family history of Alzheimer disease might obtain the genetic test, which would guide their clinician in initiating treatment. The approach would be similar to treating hypertension or hypercholesterolemia: A high cholesterol level informs the clinician to prescribe a statin drug to a patient to lower the future risk for a heart attack or stroke.

Results from the present study may help push Alzheimer disease diagnosis and treatment in the direction of prevention and help investigators discover useful prevention treatments. We know that Alzheimer disease is a heterogeneous condition, and subtypes of the condition stratified according to genetic risk profiles could be the critical step necessary to the discovery of new drug prevention treatments.



Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.