Carrying the Alzheimer risk variant ApoE4 is associated with an accelerated decline in memory function in midlife, according a 15-year longitudinal study published in yesterday’s New England Journal of Medicine. The work, from Eric Reiman from the Banner Alzheimer Institute and University of Arizona, Richard Caselli from the Mayo Clinic in Arizona, and colleagues, suggests that people with at least one ApoE4 allele start to show accelerated age-related memory defects in their fifties, compared to non-carriers. The decline may be the neuropsychological correlate of accelerated AD pathology in the carriers, the authors suggest.

The specter of faster cognitive decline, with no means of prevention, only adds to the complicated question of whether, and under what conditions, people should learn their ApoE genotype. Another paper in the same issue of NEJM can offer some reassurance on that count by showing that healthy adults handled the news of their E4 status without any major psychological upsets. That result comes from the REVEAL study, with Robert Green and Lindsay Farrer of Boston University Medical Center in Boston, Massachusetts, writing for the group. The results indicate that people (in this case, the healthy adult children of parents with AD) can accept a probabilistic risk assessment for a disease that has no treatment or cure, without experiencing high levels of anxiety, stress, or depression, at least in the short term. The long-term effects of such testing, however, and the ramifications for the wider population, remain to be seen.

A Steeper Slope
The data linking ApoE4 to age-related memory decline comes from a longitudinal study headed up by Caselli, and involving both imaging and neuropsychological testing in a group of cognitively normal adults aged 21-97 years. First author Caselli and colleagues tested 317 ApoE4 carriers (79 homozygous and 238 heterozygous) and 498 non-carriers on a battery of cognitive tests at intervals of one to two years for an average of five years. To keep the focus on aging in healthy people, any subjects that developed mild cognitive impairment or AD in the course of the study were excluded.

The results showed that between the ages of 55 and 60, the trajectory of decline in a test of long-term memory (the Auditory-Verbal Learning Test, or AVLT-LTM) began to diverge between carriers and non-carriers. In addition, there was a gene dosage effect: The E4 homozygotes dropped off the most, the heterozygotes not so much, and the non-carriers even less. The early drop was specific to the memory test, as it was not apparent with scores from tests of executive and language skills, or visuospatial function.

Finding the first evidence of differential cognitive aging in a memory domain in the 55-60 age window is significant in light of the results of recent imaging studies, Caselli told ARF. “We know from some of the imaging studies that we’ve been doing that this is an age at which we start to see amyloid deposition on PIB-PET scans (see ARF related news story on Reiman et al., 2009), and we know that from fluorodeoxyglucose uptake studies that you see metabolic abnormalities around this age as well. So there is imaging evidence for an Alzheimer-like phenomenon happening at this age, but it has never happened before that somebody has shown any cognitive ramification of that, and now we have.”

While the study does not address the mechanism of memory decline, the investigators favor the explanation that they are picking up a very early stage of AD. “One can ask the question, When does Alzheimer disease begin?” Caselli said. “No study before this has been able to show the initiation point, the inflection point at which people go from being normal to starting to be abnormal. We’ve been able to capture that, to show the inflection point in the 55- to 60-year-old group. As defined neuropsychologically at least, one could argue that that’s the beginning of AD.” That interpretation fits with data showing that ApoE4 carriers are not only at higher risk of AD, but also tend to get the disease earlier than non-carriers.

Too Much Information?
ApoE4 testing is a sticky issue because either a positive or negative result is fraught with uncertainty. As a genetic marker, ApoE4 is neither necessary nor sufficient for developing AD, which means that a person with the high risk variant may never get the illness, and conversely, one lacking the liability may still come down with the disease. On top of that, there is currently no effective treatment or reliable prevention. As author Farrer puts it, “The test has a predictive value issue, and then you have layered on top of that the question of what is the benefit of having the information.”

The REVEAL study did not aim to answer that question, but instead focused on the issue of whether it was harmful to people to know their ApoE4 status. The investigators adopted a randomized clinical trial design, starting with subjects with a familial history of AD in order to enrich their sample for ApoE4 carriers. They first genotyped 162 asymptomatic adult children of parents with AD and then randomized the subjects to two groups, one of which received their genotype information and one of which did not. Other than that, the groups went through identical counseling and testing regimes. At the end, all participants received an individually tailored report of their risk for AD based on their demographic characteristics, but only the disclosure group received an additional estimate with their ApoE genotype factored in. The investigators measured anxiety, depression, and test-related stress six weeks, six months, and one year after the results were delivered.

The results showed that there was no difference in any of the psychological measures between the groups who learned their genotype and those who did not. There was a small and just statistically significant increase in measures of distress at six weeks in people with E4 versus those without, but the increase was transient and modest. Overall, the results suggest that ApoE testing does not cause major psychological concerns. Granted, the group as a whole is not representative of the general population because they had a parent with AD and came forward voluntarily to be tested. However, Farrer says, “For this group of folks, we saw that this was psychologically a safe test.”

Farrer says that part of the objective in this study was to figure out the minimal situation where the testing could be done safely and accurately. “We do believe there needs to be some education provided to people, but it does not require a long, drawn-out protocol. We’ve shown that the process can be streamlined.”

But what about testing that is done outside of a carefully controlled study? For 10 months during 2008, ApoE4 genotyping was offered direct to consumers, through a company that has since gone out of business (see ARF related news story). (Farrer and coauthor Norman Relkin of Weill Cornell Medical College in New York were both consultants to the company, Smart Genetics). Understanding a report that comes over the Internet would require a fairly high level of sophistication in the recipient, Farrer concedes, in the absence of a genetic counselor or physician.

Though Smart Genetics is defunct, other companies offer genome scan services that include ApoE4 markers, but the problem with that is they do not provide enough genetic information to assess risk accurately, Farrer said. Such scans use single nucleotide polymorphisms to tag the ApoE risk allele, which means they can tell consumers if they have an E4 allele, but not an actual genotype. However, the totality of risk depends on whether people have one or two E4 alleles, with or without protective E2 forms, along with other factors. These details are not part of the equation with genomewide scans on offer from Navigenics, 23andMe, or deCODE Genetics, for example.

While the results indicate that learning one’s ApoE4 status does no harm in the short run in this select group of subjects, the study does not address many other issues, write Rosalind Kane and Robert Kane of the University of Minnesota in Minneapolis, in an accompanying editorial. They stress that the study looked at psychological questions, but not the financial or social effects of the testing. In addition, long-term effects loom large in such testing, but were not addressed. Finally, the question of how testing will translate to the general public, and especially people with no family history of AD, remains open.—Pat McCaffrey


  1. This is a very elegant and important study. The researchers studied 815 participants using a mixed model approach for cross-sectional and longitudinal data to assess the potential effects of ApoE ε4 (ApoE4) on longitudinal decline in memory in the absence of mild cognitive impairment or Alzheimer disease (AD). In ApoE4 carriers, longitudinal decline in memory, as assessed with the Auditory-Verbal Learning Test (AVLT-LTM), started earlier (before age 60) and showed a greater acceleration than in non-carriers. These data are consistent with earlier studies of this group showing a more rapid decline in memory in ApoE4 carriers that was correlated with reduced cerebral metabolism 5-10 years before the onset of cognitive symptoms (see Baxter et al., 2003; Caselli et al., 2004; Caselli et al., 2007; and Caselli et al., 2008). Interestingly, as presented by van der Flier during the first ApoE symposium during ICAD 2009 and published by her group earlier this year (van der Vlies et al., 2009), cognitive decline in early AD is ApoE4-dependent. In a study including 99 patients with early onset AD and 192 patients with late-onset AD, patients with early onset AD showed a more rapid cognitive decline, as assessed by the change in scores on the Mini-Mental State Examination (MMSE), and in this group, non-ApoE4 carriers showed a greater cognitive decline than non-ApoE4 carriers. In ApoE4 carriers, there was no difference in the rate of cognitive decline between early and late-onset AD.

    As presented in the second ApoE symposium during ICAD 2009, we have follow-up data on our study showing detrimental effects of ApoE4 on performance in a translational test of object recognition (Novel-Image-Novel-Location [NINL]) in 116 non-demented healthy elderly study participants (mean age 81 years) (Berteau-Pavy et al., 2007) who were invited to return six and 18 months after the baseline session. Using a longitudinal study design, effects of ApoE4 on NINL test performance were assessed in study “dropouts,” participants who did not return for the second and/or third session(s), and “finishers,” participants who returned for all sessions. While the dropout rate was higher in ApoE4 than non-ApoE4 carriers, and ApoE4 carriers with a low NINL score were more likely to drop out of the study than those with a high NINL score, in finishers ApoE4 carriers had higher NINL scores than non-ApoE4 carriers, and the scores of these ApoE4 finishers did not decline over the 18 months of the study.

    Together, these results show that although the direction of these ApoE4 effects might depend on the study population (healthy elderly without MCI, MCI, early AD, or late AD), and the cognitive measure used, these results have important implications for clinical trials and emphasize the importance to assess ApoE genotype. For example, in the promising Dimebon data from Doody et al. (see Doody et al., 2008), a different proportion of ApoE4 carriers in the two treatment groups could have either contributed to the observed effect or have masked a larger therapeutic effect. In addition, and perhaps more importantly, the treatment response to Dimebon, as seen for several other drugs, might depend on ApoE4 status as well.


    . Apolipoprotein E epsilon 4 affects new learning in cognitively normal individuals at risk for Alzheimer's disease. Neurobiol Aging. 2003 Nov;24(7):947-52. PubMed.

    . Longitudinal changes in cognition and behavior in asymptomatic carriers of the APOE e4 allele. Neurology. 2004 Jun 8;62(11):1990-5. PubMed.

    . Cognitive domain decline in healthy apolipoprotein E epsilon4 homozygotes before the diagnosis of mild cognitive impairment. Arch Neurol. 2007 Sep;64(9):1306-11. PubMed.

    . Correlating cerebral hypometabolism with future memory decline in subsequent converters to amnestic pre-mild cognitive impairment. Arch Neurol. 2008 Sep;65(9):1231-6. PubMed.

    . Most rapid cognitive decline in APOE epsilon4 negative Alzheimer's disease with early onset. Psychol Med. 2009 Nov;39(11):1907-11. PubMed.

    . Effects of sex and APOE epsilon4 on object recognition and spatial navigation in the elderly. Neuroscience. 2007 Jun 15;147(1):6-17. PubMed.

    . Effect of dimebon on cognition, activities of daily living, behaviour, and global function in patients with mild-to-moderate Alzheimer's disease: a randomised, double-blind, placebo-controlled study. Lancet. 2008 Jul 19;372(9634):207-15. PubMed.

    View all comments by Jacob Raber
  2. Caselli and colleagues show that in asymptomatic, clinically normal individuals with a mean age about 60, ApoE4 carriers have an increased mean rate of longitudinal decline on an auditory verbal learning test. Effects are ApoE4 dose-dependent. This decline, studied in a large cohort of individuals, supports the idea that a greater percentage of ApoE4 carriers are developing “preclinical” Alzheimer disease at an earlier age than non-carriers. Prior data show that amyloid deposition in the brain as reflected by amyloid imaging and a decline in CSF Aβ42 begins to occur in some ApoE4 carriers in their fifties. These data are consistent with there being subtle learning-related changes that begin to occur either about the same time or just after this pathology is beginning to build up.

    View all comments by David Holtzman
  3. A group of researchers at the Arizona Alzheimer’s Consortium led by Richard Caselli and Eric Reiman have been recruiting and following a large number of middle-aged persons who are either heterozygous or homozygous for the ApoE ε4 allele. Their current findings, based on extensive longitudinal study, make a very important observation about the relationship between carriage of the ε4 allele and cognition. While their subjects were still clinically normal, those with one or two copies of the ε4 allele experienced a decline in their delayed recall ability at an average age of 60 years. This observation is very valuable because it adds confirmatory weight to several suspicions about the relationship between ApoE ε4 carriage and Alzheimer’s disease. First, the results are consistent with the view that among the earliest clinical impacts of ApoE ε4 carriage is anterograde amnesia. This is of critical importance because it shows that ApoE accelerates the temporal profile of pathological AD. These findings support the contention that carriage of an ε4 allele lowers the age of onset of AD, on average. Furthermore, the results lend support to the use of ApoE genotype as a proxy for incipient AD.

    View all comments by David Knopman

Make a Comment

To make a comment you must login or register.


News Citations

  1. More ApoE4 Means More Amyloid in Brains of Middle-Aged
  2. News Brief: Controversial ApoE Testing Ends for Now

Paper Citations

  1. . Fibrillar amyloid-beta burden in cognitively normal people at 3 levels of genetic risk for Alzheimer's disease. Proc Natl Acad Sci U S A. 2009 Apr 21;106(16):6820-5. PubMed.

Further Reading

Primary Papers

  1. . Effect of genetic testing for risk of Alzheimer's disease. N Engl J Med. 2009 Jul 16;361(3):298-9. PubMed.
  2. . Longitudinal modeling of age-related memory decline and the APOE epsilon4 effect. N Engl J Med. 2009 Jul 16;361(3):255-63. PubMed.
  3. . Disclosure of APOE genotype for risk of Alzheimer's disease. N Engl J Med. 2009 Jul 16;361(3):245-54. PubMed.