Having one parent with Alzheimer disease is hard enough; having two must be devastating. People carrying that burden may have even more to worry about if data published this week is replicated elsewhere. In the March Archives of Neurology, researchers led by Thomas Bird at the University of Washington, Seattle, report that a person’s risk of getting late-onset Alzheimer disease (AD) shoots up if both parents had the disease. The family history of the parents themselves may also be influential, according to the study, though in a subtler way. The researchers report that while having parents with a family history of AD does not increase a child’s risk of getting the disease, it does reduce the age of onset. “The implication is that we all have risk factors for Alzheimer’s that are probably individually very small, but if you carry multiple small changes you increase your risk to a discernible difference,” Suman Jayadev, lead author on the study, told ARF.

Numerous studies have documented the prevalence of AD in the general population. The commonly accepted figure is between six and 13 percent of those over age 65, though that number steadily increases with age (some estimates suggest nearly half of those over age 85 have the disease). Given that prevalence, one might think that families with two affected parents would be hard to find. In fact, Jayadev said that only a small percentage of families registered at the University of Washington Alzheimer’s Disease Research Center are in that situation. “The reason we did this [study] is that enough people have come to us and said ‘both my parents have AD, so what is my risk?’ So there are enough people out there with two parents affected who want to know what their differential risk is,” said Jayadev. She also suggested that this may be why these findings have gotten media attention. Bird discussed this study during a radio talk show this week on Boston’s National Public Radio station WBUR, which also featured AD researchers Reisa Sperling and Pierre Tariot.

Expanding an earlier pilot study of 31 spouse pairs with AD (Bird et al., 1993), Jayadev and coworkers have now examined 111 couples with AD and their 297 surviving adult children. All told, 67 of the children (22.6 percent) were diagnosed with AD at the time of the study. High as that figure seems, it is most likely too low because 79 percent of unaffected children have not yet reached age 70. “Though we don’t know, we have a feeling that the prevalence might be higher,” said Jayadev. The researchers plan to keep following this cohort of patients, and Jayadev said more will be known in 10 years or so as these people start to get older. “When we broke things down by age of affected individuals, we see that overall there seems to be a higher incidence of AD in these children than has been published for the general public,” added Jayadev. For example, looking at the 143 study participants aged 65 and older, the researchers found that 36 percent of the children were affected, while the prevalence rose to 42 percent in those aged 70 and older.

Looking beyond the parents, the researchers found that the parents’ own history bears on their children’s age of onset. Children with affected parents having no family history had an average age of onset around 72 years. This fell to 60 years if one parent had a family history, and 57 years if both parents had a family history. This may be related to gene dose, suggest the authors.

In addition to the earlier pilot study conducted by Bird and colleagues, the Multi-Institutional Research in Alzheimer Genetic Epidemiology (MIRAGE) study also looked at the effect of having two affected parents (see Lautenschlager et al., 1996). “Though the goal there was to assess the risk of having a first-degree relative, not necessarily a parent, with AD, they had such a huge number of probands that they were able to parcel that out,” said Jayadev. The MIRAGE study found that having one affected parent endows a lifetime risk of getting AD of 37 percent, while for those with two affected parents that jumps to 54 percent. The latter figure translates to nearly five times the risk in the general population. “Though their goals were different and we had a larger number of people with two affected parents, the results are very comparable,” said Jayadev.

These familial studies show that, though small, the impact of genetic inheritance on sporadic neurodegenerative disease can be significant. A different case in point is the glucocerebrosidase gene, which, according to a second Archives of Neurology paper by Bird and colleagues this week, increases the susceptibility to Lewy body disorders in people of European ancestry. Researchers led by Cyrus Zabetian, also at Washington University, compared glucocerebrosidase (GBA) polymorphisms in 721 patients with Parkinson disease (PD), 57 people with dementia with Lewy bodies (DLB), and 554 controls. First author Ignacio Mata and colleagues found that the frequency of the two most common GBA mutations (N370S and L444P) are significantly higher in the PD and DLB populations. Though previous smaller studies have questioned the significance of these mutations, this larger study now supports the idea that GBA mutations confer risk. However, the authors point out that larger-scale study of thousands of patients will be needed to confirm this finding.

As for AD in children with two affected parents, it is not clear what the genetic influence may be. ApoE4, of course, is a well-known risk factor for AD, but it does not appear to completely explain the increased risk in these families. “There was a predominance of ApoE4 alleles, but there were also a number of patients who did not carry two, or even one, E4 allele, so we do not think it is the only factor,” said Jayadev.

Jayadev agreed that it would be worth studying genetic or biomarkers in these families. “They would be an excellent cohort of people to do that with, because we know that clinically they seem to have a higher risk, and it would be really interesting to see if whatever biomarkers bear out to be relevant may relate to their clinical presentation,” she said. The cohort might even prove useful for identifying protective factors, as it contains families where both parents have AD but none of their children do. The researchers do not presently have plans to study genetics or biomarkers.—Tom Fagan


  1. The findings in this study are consistent with an increasing number of studies indicating that family history is an important, but poorly understood risk factor for the development of Alzheimer disease. This study suggests that there are unknown genetic risk factors for Alzheimer disease which may be as important, if not more important, than the known genetic risk factor, apolipoprotein (ApoE) 4. The high prevalence of ApoE4 in the family history group also raises the possibility that the unknown family history factor may have confounded prior studies, which did not take family history into account. In other words, prior research that made comparisons between research groups defined by ApoE genotype may have been biased by the high prevalence of the unknown family history genetic factor in the ApoE4 group.

    View all comments by Mark Sager
  2. Very interesting study, though not quite conclusive. Research designs and analyses could potentially have been made even better by, for instance, including relevant controls. Hopefully, the potential presence of recessive FAD cases or gene-gene interactions might be established. It would be interesting to examine whether Hiroshi Mori's new APP mutation can be identified in any of the patients (Tomiyama et al., 2008).


    . A new amyloid beta variant favoring oligomerization in Alzheimer's-type dementia. Ann Neurol. 2008 Mar;63(3):377-87. PubMed.

    View all comments by Takaomi Saido
  3. Important as is the paper by Jadaev et al., and going a long way to confirm the genetic etiology of "ordinary" AD as distinct from obviously familial EOAD, one is surprised to find to it, with all those percentages, a "biometrician" flavor (the Galton-Pearson side of the famous early twentieth-century quarrel with the "Mendelians" or "geneticists" led by Bateson). Using "censored data" corrections to account for the people who, either through premature death or by observation cut-out, were not observed beyond the estimated range of AD manifestation in the parents, it should have been possible to estimate a Mendelian segregation ratio, as did Breitner et al. 20 years ago. It was one half with one first-degree affected relative in the Breitner paper; it would be three quarters in the present case—both parents affected—under the same Mendelian transmission. The very high theoretical segregation ratio would explain the fact that a high proportion of affected parents were observed by Jadaev et al. in spite of the severe limitation of the observation span. Indeed, we showed (Bruni et al.) that monogenic dominant Mendelian transmission on the model of familial EOAD is compatible with the epidemiology of apparently sporadic LOAD under the simple hypothesis of a theoretical mean age of expression ~80 years of the mutated gene(s), if stochastic repartition (standard deviation ~6 years) of the age of expression, and data censorship are accounted for.


    . Age-dependent expression of familial risk in Alzheimer's disease. Am J Epidemiol. 1988 Sep;128(3):536-48. PubMed.

    . Alzheimer's disease: a model from the quantitative study of a large kindred. J Geriatr Psychiatry Neurol. 1992 Jul-Sep;5(3):126-31. PubMed.

    View all comments by Jean-François Foncin
  4. This paper adds momentum to the growing concept that GBA mutations play a role in the susceptibility to the Lewy body disorders PD and DLB. Importantly, a large number of patients and controls were studied, although only the two most common GBA mutations were searched for. A small but significant rate of GBA mutations was detected in the PD group. Similar results have been documented in a number of other locations and ethnicities. Presumably in these GBA-mutated PD patients, α-synuclein processing is altered, and so the key question for future studies is to understand how glucocerebrosidase mutations lead to processing changes in α-synuclein. Potential influencers include an increase in glucosylceramide substrate accumulation (after all, synuclein is known to be a lipid-binding protein), GBA misfolding, and trafficking failure and/or general lysosomal dysfunction. Detailed molecular analysis will be needed to unravel this intriguing puzzle.

    View all comments by Valerie Cullen

Make a Comment

To make a comment you must login or register.


Paper Citations

  1. . Conjugal Alzheimer's disease: is there an increased risk in offspring?. Ann Neurol. 1993 Sep;34(3):396-9. PubMed.
  2. . Risk of dementia among relatives of Alzheimer's disease patients in the MIRAGE study: What is in store for the oldest old?. Neurology. 1996 Mar;46(3):641-50. PubMed.

External Citations

  1. radio talk show

Further Reading

No Available Further Reading

Primary Papers

  1. . Conjugal Alzheimer disease: risk in children when both parents have Alzheimer disease. Arch Neurol. 2008 Mar;65(3):373-8. PubMed.
  2. . Glucocerebrosidase gene mutations: a risk factor for Lewy body disorders. Arch Neurol. 2008 Mar;65(3):379-82. PubMed.