Medscape www.medscape.com
To Print: Click your browser's PRINT button.
NOTE: To view the article with Web enhancements, go to:
http://www.medscape.com/viewarticle/484814

Conference Report
Unraveling the Origin of Alzheimer's Disease
9th International Conference on Alzheimer's Disease and Related Disorders; July 17-22, 2004; Philadelphia, Pennsylvania

Sara M. Mariani, MD, PhD

Medscape Molecular Medicine 6(2), 2004. © 2004 Medscape

Posted 09/08/2004

The harp that once through Tara's halls
The soul of music shed
Now hangs mute on Tara's walls
As if that soul were fled

Thomas Moore

Introduction

What would happen if 15 million people were affected by Alzheimer's disease (AD)?

Far from being the product of a fertile imagination, this is the number of cases currently projected for the year 2050. Today, about 4.5 million people are diagnosed with AD. The direct and indirect costs, now calculated at approximately $100 billion in the United States and 87 billion Euros in Europe, would skyrocket proportionally.[1-4]

If true, this scenario presented by leading investigators and epidemiologists at the 9th International Conference on Alzheimer's Disease and Related Disorders (ICAD) is just short of a public health emergency for developed countries with aging populations.

Genetic, behavioral, and environmental risk factors all appear to contribute to the multifactorial etiopathogenesis of AD[5-7]:

  • Age;

  • Genetic factors, with a 40% concordance between identical twins[6,8];

  • APOE4 genotype: increased risk but only incomplete association;

  • Ethnicity: Latinos appear to develop AD almost 7 years earlier on average than non-Latino Americans; African Americans show a 2- to 3-fold higher incidence rate, and thus higher prevalence[9,10];

  • Cardiovascular risk factors: heart disease, high cholesterol levels, high blood pressure, and stroke (hypooxygenation of the brain); and

  • Lower education (even among subjects with very similar genetic predispositions to intellectual achievement, such as identical twins).[11]

What to do? The Alzheimer's Association is calling for a campaign called "Maintain your brain" aimed at introducing changes in lifestyle that can help to reduce, whenever possible, the prevalence of these risk factors and thus delay or prevent the onset of AD.[12] Among them:

  • Weight loss and healthy food diet;

  • Reduction of cholesterol levels and high blood pressure; and

  • Complex leisure activities with physical, mental, and social interactivity components.


Epidemiology of AD

Although there is no association between AD and sex (men and women are equally affected), age represents a very strong risk factor. According to the results of a prospective study, AD rates of 2.8 per 1000 person-years in the age group 65-69 rise to 56.1 per 1000 person-years in subjects older than 90 years of age. The relative increase is highest in the 75-84-year age group.[13]

A higher educational level (> 15 years vs < 12 years of education) is associated with a decreased risk of AD; however, the association is also dependent on the baseline cognitive screening test score.[14]

As noted by Dr. Lenore Launer,[15] of the National Institute on Aging, Bethesda, Maryland, a study of the trajectory of AD, from a relatively healthy status to minimal cognitive impairment (MCI) on to full-blown AD, is of primary importance to understand the risk factors that favor the development of AD. In this trajectory from baseline to initial dementia, also factors linked to age-related changes, comorbidities and mortality play a significant, if not critical, role in determining who will develop AD.[16-18]

We have certainly improved on our knowledge of AD and in our diagnostic abilities, if we look back at a statement published in a 1981 issue of BMJ noting that an AD diagnosis was often put forth ""when the patient was older and less intelligent than the doctor."

Standard diagnostic guidelines have been established; multidisciplinary strategies are now prevailing in diagnosis and treatment; new technologies, particularly imaging, are available to refine diagnosis; and community-based approaches are being implemented for the management of patients with AD. De novo cohorts are also being established to replace extant ones and to evaluate whether we are witnessing changes in risk factors for AD to which the current general population is being exposed.[15]

A number of paradoxes and inconsistencies are reported in the literature on the epidemiology of AD, as noted by Dr. Launer, and it is important to address them to understand how many variables may affect epidemiologic analysis in this field.

For example, increased blood pressure was expected to be associated with poor cognitive function in older individuals. Some studies have, however, reported an inverse association: Very high blood pressure was found to play an apparently protective role from AD, and a very low blood pressure to predispose to AD. In prospective association studies, the increased risk for very low blood pressure appeared to be approximately 4.7.

Similarly, increased cholesterol levels were expected to be associated with an increased risk of dementia. Contradictory results have been obtained in 3 different prospective studies: an increased risk in the Kuopio cohort (relative risk [RR], 2.2), no association of AD with high cholesterol levels in the Framingham cohort (RR, 0.95), and protection in a third cohort (RR, 0.55).[15,16]

In a cohort of Japanese American men born between 1900 and 1919, studied in Honolulu, Hawaii, since 1965, it was, however, found that their systolic blood pressure increased at about 59 years of age but declined at later times. Their diastolic blood pressure, on the other hand, underwent a progressive reduction in these subjects after reaching a peak at 60 years of age. Thus, what emerged from this study was that risk factors may substantially change with age, and they may be measured at a time when their absolute values have already changed, leading to incorrect or incomplete assessment of their effect and relevance in the etiopathogenesis of AD.[15]

In the second example, cholesterol levels are known to drop in the presence of comorbidities, such as cardiovascular disease, cancer, and chronic obstructive pulmonary disease, thus potentially altering the epidemiologic findings and our conclusions as to their importance, when measured at the incorrect time points.

One more source of complexity in the epidemiologic analysis of AD is an involuntary oversight of cause-effect relationships. Attention has to be paid to the factors that are indeed associated with risk, but that are not the consequence of the dementing process, as seen, for example, with the extent of social activity. Older people with a consistently low social activity have a RR of 1.35 to develop AD vs a RR of 0.94 for those that have a consistently high social involvement. Of note, however, the risk apparently increased to 1.75 in older people who changed their habits and drastically reduced their social activities. In this case, however, cognitive impairments may have already been present, thus inducing changes in the behavior of the person affected. Drastic reductions in social interactions may have been more a consequence than a risk factor for AD in some individuals not correctly diagnosed.[15]

One more paradox emerged from correlations between smoking and AD in men of the Honolulu cohort. The RR of AD increased from 1 in never-smokers to 2 and 2.5 in light and heavy smokers, respectively. Of note, however, the RR of AD was only 1.5 in very heavy smokers. Overall differences in early mortality rates from other diseases or accidents were not enough to account for this apparent paradox.

Genetic analysis of the APOE alleles, however, yielded an explanation for these apparently odd findings. Smokers with the APOE3 allele had a better survival than smokers with the APOE4 allele, who had the lowest survival. And the APOE4 allele is known to be an independent factor that substantially increases the risk for AD. Thus, risk factors for AD may even subtly modify the composition of the cohorts analyzed when modifying risk or outcomes of other diseases or conditions.[15]

A number of risk factors are being assessed and reassessed for their relative contribution to the development of AD, including physiological factors, cholesterol levels, insulin levels, blood pressure, inflammation, homocysteine levels, hormonal levels, and the presence of obesity or diabetes. Most of them are known to contribute to peripheral arterial disease, coronary heart disease, and stroke, and their role in AD is being further evaluated in ongoing, prospective cohort studies.[16]


Secular Changes

Dr. Launer concluded her presentation with a provoking thought. The past few decades have seen extensive changes in public awareness of health, more prevention campaigns, and countless successes achieved by medical and surgical treatments in diagnosed patients. As a consequence of all these efforts, we may be witnessing secular changes in risk factors among the population of wealthier countries, changes that may affect the long-term prevalence and incidence of risk factors, 30-50 years down the road.[15]

We have, for example, seen a secular change in blood pressure levels with a shift to the left of the curves, leading to high levels at a younger age than in the past. We are also seeing a substantial reduction in the numbers of subjects with uncontrolled hypertension, of the total number of smokers, and an increase in the number of individuals who keep low cholesterol levels through the use of lipid-lowering drugs or suitably designed diets and physical activity. On the other hand, we are witnessing, particularly in the United States, a rather worrying increase in the incidence and prevalence rates of obesity and type 2 diabetes, even in very young subjects.

As the populations change, risk factors may change, and epidemiologists will try to keep pace. Prospective studies now under way will further address the relevance of some of the results obtained so far. New cohorts will, however, have to be established to identify new emerging risk factors that may significantly change the way in which we look at the cofactors favoring a high incidence rate of AD or an unfavorable early onset.


Genetics of AD

As seen from its complex epidemiology, AD is a multifactorial disease in which genetic, environmental, and stochastic factors interact to induce neuronal damage and dementia. From a genetic point of view, AD presents multiple patterns of inheritance, with single mutated genes, modified genes, and multigenic patterns involved in its ethiopathogenesis.[19-21]

As illustrated by Dr. Gerard Schellenberg,[22] of the University of Washington, Seattle, inherited mutations of the alpha-beta-gamma secretase complex are associated with a dominant pattern of AD inheritance and early onset of the disease (40-60 years of age). AD, in these cases, is fully penetrant (all individuals carrying one of such mutations develop AD), leading to accumulation of excess amounts of amyloid-beta (Abeta) and its fragment Abeta-42, in addition to a fibrinogenic mutant of Abeta. A total of 8 mutations have been identified so far, in 8 different gene locations.

Mutations in the presenilin (PS)1 and 2 genes lead to the typical neuropathologic changes described as "cotton wool" plaques. AD is transmitted in these subjects in an autosomic dominant fashion. Mutations in PS1 may lead to a very early AD onset (28-65 years of age) with full penetrance. Mutations in PS2 may manifest later in life with the development of AD at 43 to > 80 years of age. Approximately 80 mutations have now been found scattered in the PS1 and PS2 genes. Because these gene products contribute to the formation of the secretase complex, when mutated they induce a shift in the gamma-site cleavage with the generation of a larger form of Abeta. A third category of single-gene mutations affects the tau protein and is associated with different but characteristic neuropathologic findings.[23-25]


Late-Onset AD Genes

Because cognitive function undergoes a progressive deterioration before the appearance of full-blown AD, many efforts are under way to identify the genes associated with intermediate disease stages, such as MCI. In addition, as diagnostic efforts are being targeted at earlier diagnoses, genetic studies are trying to identify the genes that may affect the length of the prodromic phase and duration of the MCI stage, and play a significant role in the progression to AD.[18]

Apolipoprotein E (ApoE) is the only gene associated, so far, with late-onset AD, and its progress in AD research has been accompanied by a few controversies. Three polymorphisms, APOE2, APOE3, and APOE4, have been described in exon 4, all of which give rise to amino acid changes. When AD patients were typed for this locus, those individuals carrying 2 E4 alleles (homozygous) had AD onset at a significantly younger age than patients homozygous for the E3 allele. Patients with the E2 and E3 alleles were the last to develop AD. No information was obtained for the E2/E2 haplotype, as it was absent in this cohort of patients. A similar APOE-related effect on the number of years of deviation from average AD onset was found in patients carrying PS1/2 mutations.[23-25]

The higher risk of an earlier AD onset conferred by the E4 allele seems, however, to be still modified by environmental factors. Urban dwellers with the E4/E4 phenotype, in fact, appeared to have a later onset than E4/E4 individuals living in more isolated, rural settings. Also, polymorphisms in the promoter region of the APOE gene appeared to influence the onset of AD. Different alleles located in the promoter and intronic regions seemed to have different effects in different ethnic populations, although their contribution to AD has yet to be fully explored. Additional multiple tissue enhancers and brain control regions may add another level of complexity to the control of APOE gene expression.[22]

Do other genes influence the late onset of AD? In Dr. Schallenberg's estimate, there may be 4-6 quantitative trait loci involved, with at least 3 or 4 genes of which the influence is as strong or stronger than that seen with the APOE gene. Two different methodologies may be used to identify these new genes. One relies on linkage analysis in families with late-onset AD, chromosomal localization, positional cloning, and gene identification that, ultimately, provide specific genetic information for association studies. The other strategy envisions the selection of predefined genes on the basis of their pathobiological functions and analysis in case-control studies of families with late-onset AD.

A number of loci have now been identified that may potentially affect the development and progression of AD, although the strength of their associations varies in different studies[22,26,27]:

19q13 << 10q22 >
9p21 > 12p13.3 > Xp11
1.3   3.9   1.8   1.3   1.2   RR
(multiple nonparametric score)

Two other studies have highlighted the 19q13 and 10q22 linkage:

19q13 >> 9p22.1 > 7q31 > 4q32
5.69   2.97   1.56   1.30   RR

9p21 > 12q21 > 2p21 > 10q22

19p > 19q13 > 10q22 > 9

Ongoing association studies have a high power to detect a pathogenic allele in large case-control studies, and they are relatively easy to perform with at least 50 cases and 50 age-matched controls. Although they offer the possibility to test all genes of interest simultaneously, they can be associated with a high rate of false positives, a quite discouraging result. The false positives may be related to multiple testing, a limited knowledge of pathogenesis and gene function, the use of mixed populations, and a linkage disequilibrium effect. In more than 1 occasion, a lack of replication by independent groups was due to the analysis of cohorts with substantially different compositions.[22]

Some of the limitations are known and the challenges are many. Nonetheless, many efforts are under way to unravel the genetic basis of AD. The concomitant identification of those behavioral and stochastic factors associated with protection or increased risk of AD is also expected to provide critical clues to devise new strategies for the prevention and management of AD.

References

  1. Doraiswamy PM. Marked increase in Alzheimer's disease identified in Medicare claim records between 1991 and 1999. Program and abstracts of the 9th International Conference on Alzheimer's Disease and Related Disorders; July 17-22, 2004; Philadelphia, Pennsylvania.
  2. Kulasingam S. Costs associated with increasing severity of dementia -- a review of the literature over 15 years. Program and abstracts of the 9th International Conference on Alzheimer's Disease and Related Disorders; July 17-22, 2004; Philadelphia, Pennsylvania.
  3. Zbrozek A. Costs of formal dementia care in Europe. Program and abstracts of the 9th International Conference on Alzheimer's Disease and Related Disorders; July 17-22, 2004; Philadelphia, Pennsylvania.
  4. Bloom BS, de Pouvourville N, Straus WL. Cost of illness of Alzheimer's disease: how useful are current estimates? Gerontologist. 2003;43:158-164.
  5. Haan MN, Wallace R. Can dementia be prevented? Brain aging in a population-based context. Annu Rev Public Health. 2004;25:1-24. Abstract
  6. Ritchie K, Lovestone S. The dementias. Lancet. 2002;360:1759-1766. Abstract
  7. Cummings JL, Cole G. Alzheimer disease. JAMA. 2002;287:2335-2338. Abstract
  8. Plassman B. Alzheimer's disease in the NAS-NRC Twin Registry of WWII veterans. Program and abstracts of the 9th International Conference on Alzheimer's Disease and Related Disorders; July 17-22, 2004; Philadelphia, Pennsylvania.
  9. Laditka J. Epidemiology of Alzheimer's disease: race effects, area variation and clustering. Program and abstracts of the 9th International Conference on Alzheimer's Disease and Related Disorders; July 17-22, 2004; Philadelphia, Pennsylvania.
  10. Clark C. Latino patients with AD have an earlier age of symptom onset compared with Anglos. Program and abstracts of the 9th International Conference on Alzheimer's Disease and Related Disorders; July 17-22, 2004; Philadelphia, Pennsylvania.
  11. Gatz M. Genetic effects do not account for the relationship between education and dementia. Program and abstracts of the 9th International Conference on Alzheimer's Disease and Related Disorders; July 17-22, 2004; Philadelphia, Pennsylvania.
  12. Pope SK, Shue VM, Beck C. Will a healthy lifestyle help prevent Alzheimer's disease? Annu Rev Public Health. 2003;24:111-132.
  13. Kukull WA, Bowen JD. Dementia epidemiology. Med Clin North Am. 2002;86:573-590. Abstract
  14. Kukull WA, Higdon R, Bowen JD, et al. Dementia and Alzheimer disease incidence: a prospective cohort study. Arch Neurol. 2002;59:1737-1746. Abstract
  15. Launer L. Epidemiology of Alzheimer's disease: lessons from cardiovascular studies. Program and abstracts of the 9th International Conference on Alzheimer's Disease and Related Disorders; July 17-22, 2004; Philadelphia, Pennsylvania.
  16. Launer LJ. Demonstrating the case that AD is a vascular disease: epidemiologic evidence. Ageing Res Rev. 2002;1:61-77. Abstract
  17. Mayeux R. Epidemiology of neurodegeneration. Annu Rev Neurosci. 2003;26:81-104. Abstract
  18. Luis CA, Loewenstein DA, Acevedo A, Barker WW, Duara R. Mild cognitive impairment: directions for future research. Neurology. 2003;61:438-444. Abstract
  19. Pastor P, Goate AM. Molecular genetics of Alzheimer's disease. Curr Psychiatry Rep. 2004;6:125-133. Abstract
  20. Bertoli-Avella AM, Oostra BA, Heutink P. Chasing genes in Alzheimer's and Parkinson's disease. Hum Genet. 2004;114:413-438. Abstract
  21. Rogaeva E. The solved and unsolved mysteries of the genetics of early-onset Alzheimer's disease. Neuromolecular Med. 2002;2:1-10. Abstract
  22. Schellenberg G. Genetics of Alzheimer's disease. Program and abstracts of the 9th International Conference on Alzheimer's Disease and Related Disorders; July 17-22, 2004; Philadelphia, Pennsylvania.
  23. Hyman BT, Gomez-Isla T, West H, et al. Clinical and neuropathological correlates of apolipoprotein E genotype in Alzheimer's disease. Window on molecular epidemiology. Ann N Y Acad Sci. 1996;777:158-165. Abstract
  24. Mahley RW, Nathan BP, Pitas RE. Apolipoprotein E. Structure, function, and possible roles in Alzheimer's disease. Ann N Y Acad Sci. 1996;777:139-145. Abstract
  25. Farrer LA, Cupples LA, Haines JL, et al. Effects of age, sex, and ethnicity on the association between apolipoprotein E genotype and Alzheimer disease. A meta-analysis. APOE and Alzheimer Disease Meta Analysis Consortium. JAMA. 1997;278:1349-1356. Abstract
  26. Kennedy JL, Farrer LA, Andreasen NC, Mayeux R, St George-Hyslop P. The genetics of adult-onset neuropsychiatric disease: complexities and conundra? Science. 2003;302:822-826.
  27. Wijsman EM, Daw EW, Yu CE, et al. Evidence for a novel late-onset Alzheimer disease locus on chromosome 19p13.2. Am J Hum Genet. 2004;75:398-409. Epub 2004 Jul 08.


Sara M. Mariani, MD, PhD, Deputy Editor, Medscape General Medicine; Site Editor/Program Director, Medscape Molecular Medicine


Disclosure: Sara M. Mariani, MD, PhD, has no significant financial interests or relationships to disclose.