|Home | About | Journals | Submit | Contact Us | Français|
Identifying a biomarker for Alzheimer's disease that can be obtained from a blood sample has been a goal of researchers for many years. Over the past few years a number of investigators have studied several plasma biomarkers but most frequently plasma amyloid Aβ40 and Aβ42 while others have explored the use of genetic variants as biomarkers for diagnosis or risk. This review considers the cross sectional and longitudinal data regarding plasma Aβ40 and Aβ42 as diagnostic biomarkers as well as risk biomarkers. Review of recent genome wide association studies indicates as many as 10 genetic variants have been associated with susceptibility to AD. Further analysis suggests that these factors have modest effects on risk and are thus not helpful, as yet in the diagnosis of disease. Until the function of these genes is understood, their role in risk and diagnosis will remain uncertain. Thus, there are several types of peripheral biomarkers under investigation, but more work is required before that can be deemed clinically useful.
Alzheimer's disease (AD) is among the most frequently encountered diseases in aging societies with an estimated five million people in the United States and 17 million people worldwide suffering from the disease. It is expected that these numbers will quadruple by the year 2040, by which 1 out of 45 Americans will be affected, leading to a considerable public health burden.
To date, there are no definitive diagnostic tests or biological risk markers of the disease. The diagnosis of AD during life is based on clinical examination using the criteria of the National Institute of Neurological and Communicative Disorders and Stroke (NINCDS) and the Alzheimer's Disease and Related Disorders Association (ADRDA) Work Group(McKhann, Drachman et al. 1984). Although these criteria have been found to be reliable and valid any measure that would increase diagnostic sensitivity and specificity would be highly valuable for improving early detection and intervention.
There have been approximately 26 investigations assessing plasma Aβ40 and Aβ42 as a diagnostic or as a biological risk factor (Tables 1–3). Studies of high risk populations (Table 1) have been consistent in showing elevated plasma Aβ42 levels in individuals from families multiply affected by Alzheimer's disease (AD) (early onset and late onset). Scheuner et al(Scheuner, Eckman et al. 1996) first reported elevated levels among symptomatic carriers of presenilin mutations. Recently Ringman et al(Ringman, Younkin et al. 2008)found that Aβ42 and Aβ42/Aβ40 ratio levels were elevated in unaffected familial AD mutation carriers compared with unaffected individuals with familial AD without mutations. However, Aβ42 levels were lower in mutation carriers with incipient AD characterized as having a clinical dementia rating (CDR(Hughes, Berg et al. 1982))= 0.5, supporting the hypothesis that Aβ42 decreases prior to overt disease. First degree relatives of patients with LOAD without known mutations or genetic variants have also been found to have increased plasma Aβ42(Ertekin-Taner, Younkin et al. 2008). Adults with Down syndrome who produce more plasma Aβ42 due to trisomy and triplication of the gene for APP also show high plasma levels of Aβ42 before onset of dementia(Schupf, Patel et al. 2007). Among studies of sporadic LOAD, there has been little consistency in the findings.
Cross sectional studies (Table 2) comparing patients with clinically diagnosed late onset Alzheimer's disease (LOAD) with controls showed no differences in either plasma Aβ40 or Aβ42. However, two studies that included patients with mild cognitive impairment (MCI) found high plasma Aβ42, but not Aβ40. Prospective studies (Table 3) have generally shown that higher plasma Aβ42 levels are associated with increased risk of LOAD or cognitive decline. Most of the studies cited in table 3 also indicate that the elevation in levels of plasma Aβ42 are present before or just at the onset of clinically diagnosed disease. Others have included repeated measurements of plasma Aβ42, finding that levels decreased as disease progressed. Not all studies agree. At least two prospective studies examining plasma Aβ42 showed no difference in patients with MCI who progressed to LOAD(Hansson, Zetterberg et al. 2006; Hansson, Zetterberg et al. 2007). Two others showed that low plasma Aβ42 or a low ratio of plasma Aβ42/Aβ40 was associated with more rapid cognitive decline or with disease(Graff-Radford, Crook et al. 2007; Locascio, Fukumoto et al. 2008). These findings are consistent with our hypothesis that high plasma Aβ42 is an antecedent indicator of risk for LOAD and declines with onset and progression of dementia. One explanation for the differences in the results is the time of sampling. A decline in plasma Aβ42 may already have begun when patients develop signs of MCI. Alternatively, the methods used to analyze plasma Aβ42 and Aβ40 may explain some of the differences in results, though at least two studies used the same methods for measurement of plasma Aβ with different results(Mayeux, Honig et al. 2003; van Oijen, Hofman et al. 2006). Despite the variability in levels before disease onset, there is a general consensus that plasma Aβ42 levels, and perhaps plasma Aβ40 levels as well, decrease with disease progression.
Compared to asymptomatic individuals with low plasma Aβ42 levels at initial assessment, Schupf et al(Schupf, Tang et al. 2008)reported that those with high Aβ42 levels had more than a three-fold increased risk of developing LOAD over an average of four and a half years. At the follow-up assessment when blood sampling was repeated, a decrease in plasma Aβ42 levels, but not Aβ40 levels, was related to the development of LOAD. The likelihood of having converted to LOAD 18 to 24 months before the second blood draw was three times higher when plasma Aβ42 levels had decreased by more than a half a standard deviation or when the plasma Aβ42/Aβ40 ratio decreased by more than a half a standard deviation. Thus, over time, decreasing levels of plasma Aβ42 or a decline in the Aβ42/Aβ40 ratio are sensitive indicators of recent conversion to LOAD. The authors have posited that the decline in plasma Aβ42 reflects compartmentalization and deposition of Aβ peptides in brain.
These results confirmed and extended previous findings among nondemented individuals who subsequently developed LOAD(Mayeux, Tang et al. 1999; Mayeux, Honig et al. 2003; Schupf, Tang et al. 2008) and are consistent with studies in women with MCI and among asymptomatic first-degree relatives of patients with(Assini, Cammarata et al. 2004; Ertekin-Taner, Younkin et al. 2008), groups at high risk of developing LOAD. However, these results are in distinct contrast to others that report 1) no relation between plasma Aβ peptide levels and risk of LOAD(Fukumoto, Tennis et al. 2003), 2) an association between low plasma Aβ40 and LOAD(van Oijen, Hofman et al. 2006; Sundelof, Giedraitis et al. 2008) or 3) a relation between a low plasma Aβ42/Aβ40 ratio and subsequent cognitive impairment or LOAD(Graff-Radford, Lucas et al. 2003; Graff-Radford, Crook et al. 2007). A number of factors may account for these inconsistencies. One important factor is likely to be the timing of sample collection in relation to the preclinical period or to stage of disease onset and progression. Few studies have examined risk associated with change in plasma Aβ peptide levels or change in Aβ42/Aβ40 ratio over time. In the study by Schupf et al(Schupf, Tang et al. 2008), conversion to LOAD was strongly related to a decline in Aβ42 levels and in the Aβ42/Aβ40 ratio. Similarly, in a study of healthy nondemented elderly individuals, higher initial Aβ42 levels and greater reductions in Aβ42 levels over an approximately 4 year period were associated with greater cognitive decline(Pomara, Willoughby et al. 2005). In the CSF, low levels of Aβ42 and Aβ42/Aβ40 ratios in patients with MCI are associated with higher brain amyloid load(Fagan, Mintun et al. 2006; Fagan, Roe et al. 2007)and predict conversion to LOAD(Blennow and Hampel 2003; Blennow and Vanmechelen 2003; Hansson, Zetterberg et al. 2006). This suggests that a decline in Aβ42 levels and in Aβ42/Aβ40 ratios can herald the onset of LOAD, possibly reflecting sequestration of Aβ42 in senile plaques or the formation of semi-soluble oligomers(Lesne, Koh et al. 2006; Lesne, Kotilinek et al. 2008).
Offspring of patients with AD are more likely to have lower mean plasma APOE levels compared with offspring of controls. Individuals with one or more APOEε4 alleles have lower APOE levels compared to those with other APOE alleles(van Vliet, Westendorp et al. 2009). Blood levels of progranulin are useful as an indicator of progranulin-related frontotemporal lobar degeneration, both null mutations and missense mutations as well as blood levels have also been observed in patients with Alzheimer's disease(Sleegers, Brouwers et al. 2010). Increased plasma concentration of clusterin was predictive of greater fibrillar amyloid-beta burden in the medial temporal lobe at autopsy but has not been investigated as a diagnostic or risk biomarker(Thambisetty, Simmons et al. 2010; Thambisetty, An et al. 2011).
There is insufficient evidence to permit a conclusion regarding the use of plasma Aβ40, Aβ42 or the ratio of Aβ42/Aβ40 in the diagnosis or assessment of risk of AD using cross-sectional or single measurements. There is suggestive evidence that changes in plasma Aβ40, Aβ42 or the ratio of Aβ42/Aβ40 may be associated with -- and therefore useful in identifying -- individuals at risk for developing AD. Standardization of the measurement Aβ40 and Aβ42 is required to determine whether or not plasma measurement will be useful in risk assessment of Alzheimer's disease.
There is insufficient data to permit a conclusion regarding the use the use of plasma or serum levels of APOE, clusterin or progranulin in the assessment of risk or as a diagnostic biomarker in Alzheimer's disease.
Predisposing genetic variants could, in the future, be used to predict who will develop dementia. Individuals genetically predisposed to dementia may benefit from therapeutic interventions in the early stages of the disease. Early intervention could significantly prevent or delay the onset, which in turn would improve quality of life of the patient and their relatives and would significantly reduce the public health burden.
The majority of familial early onset Alzheimer's disease (EOAD) is caused by mutations in APP, PSEN1 and PSEN2. These genes have almost complete penetrance (>85%) and a clear autosomal dominant pattern of inheritance. Thus, the presence of a mutation is virtually synonymous with disease. Complicating the wide-spread use of genetic testing for familial EOAD is the relative rarity of cases and the fact that there are multiple mutations in these genes producing the same phenotype. Thus, genetic testing requires full sequencing of the gene.
Current knowledge suggests that a variety of mechanisms underlie the various neuropathological and clinical changes, and that these have different genetic and environmental components. Thus, it is likely that late-onset cognitive impairment is a complex genetic disorder characterized by an interaction of multiple genes and the environment leading to genetic variants that have incomplete penetrance and a low-magnitude associated risk. Consistent with this notion is the fact that to date only APOE has been firmly identified as a genetic risk factor, although segregation analyses conducted in families of patients with LOAD support the presence of additional(Daw, Payami et al. 2000). With a population attributable risk that is estimated at 20–50%, the APOEε4 allele increases risk of cognitive impairment, LOAD, and ageof onset of cognitive impairment in a dose-dependent fashion: one ε4 allele is associated with a 2–3 fold increased risk, having two copies is associated with a 5–10 fold increase. Similar effect sizes have been observed for progression of cognitive impairment to dementia.
SORL1 was identified as a candidate gene in which two haplotypes in the 3' and 5' regions of the gene (Rogaeva, Meng et al. 2007)found to be associated with LOAD, with effect sizes ranging from odds ratios of 1.4 to 2.2. Subsequent studies have confirmed the association(Reitz, Cheng et al. 2011). Variants in APOE and SORL1 lack sufficient sensitivity and specificity to be used as diagnostics. Variants in both increase risk of cognitive impairment in a non-Mendelian fashion, are not fully penetrant, and that they are neither necessary nor sufficient by themselves to cause impairment. The same is likely to be true for the remaining yet-to-be-identified genetic factors associated with cognitive decline.
There have been approximately 16 genome-wide association studies (GWAS) published to date. Most of these studies have included unrelated patients with LOAD and controls, but there are three studies that used family data(Bertram, Lange et al. 2008; Poduslo, Huang et al. 2009; Wijsman, Pankratz et al. 2011). Most GWAS have confirmed the APOE association with LOAD. Two large studies have identified variants in Clusterin (CLU), phosphatidylinositol binding clathrin assembly protein (PICALM) and complement component (3b/4b) receptor 1 (CR1) as being associated with LOAD risk(Harold, Abraham et al. 2009; Lambert, Heath et al. 2009). CLU, or apolipoprotein J (APOJ), is a lipoprotein found to be part of amyloid plaques. CLU specifically binds soluble Aβ in cerebrospinal fluid to form complexes able to cross the blood brain barrier. It has also been noted that reduced levels of APOE and increased levels of CLU are correlated with the number of ε4 alleles, suggesting a compensatory induction of CLU in the brain of LOAD individuals with the APOE ε4 allele and low brain levels of APOE. CR1 encodes the complement C3b protein, which is likely to contribute to Aβ clearance(Wyss-Coray, Yan et al. 2002). PICALM is involved in clathrin-mediated endocytosis (CME), an essential step in the intracellular trafficking of proteins and lipids such as nutrients, growth factors and neurotransmitters. Of relevance to LOAD, PICALM also appears to be involved in directing the trafficking of VAMP2, which is a soluble N-ethylmaleimide-sensitive factor attachment protein receptor (SNARE) protein that has a prominent role in the fusion of synaptic vesicles to the presynaptic membrane during neurotransmitter release, a process that is crucial to neuronal function. The gene encoding bridging integrator 1 (BIN1) was also noted initially at a lower threshold of significance(Harold, Abraham et al. 2009)with stronger results in a recent replication analysis(Seshadri, Fitzpatrick et al. 2010). BIN1, a gene expressed in the central nervous system, and is reported to activate a caspase-independent apoptotic process.
Genome-wide significant results at MS4A4A on chromosome 11, CD2AP on chromosome 6, EPHA1 on chromosome 7, and CD33 on chromosome 19 were observed in independent meta-analyses by the Alzheimer's Disease Genetic Consortium and another European-American Consortium (Hollingworth, Harold et al. 2011; Naj, Jun et al. 2011). The true genetic effector for these genes has not been identified and their roles in the pathogenesis are unknown.
Can any of these genes be used as biomarkers for diagnosis or as measures of disease risk? At this point, only mutations in genes associated with early onset familial Alzheimer's disease could be considered diagnostic. These mutations are virtually 100% penentrant, and therefore those who carry a mutation will develop disease. Of interest, some patients with these mutations also have elevated plasma levels of Aβ42.
Genetic variants associated with late-onset Alzheimer's disease are not useful as diagnostic measures. For example, APOE-ε4 is strongly associated with disease risk, but it has been shown to provide little help in the diagnosis. While there are many candidate genes in which variants are, or may prove to be, associated with increased risk of subsequent disease, none are currently used as biomarkers of risk.
There is sufficient direct evidence to support the utility of pathogenic mutations identified in PSEN1, PSEN2 and APP in the diagnosis of Alzheimer's disease. There is sufficient evidence to support the use of variation at the APOE locus in the assessment of risk of Alzheimer's disease, but insufficient evidence to permit a conclusion regarding the use of variation at the APOE locus for diagnosis. There is insufficient evidence to permit a conclusion regarding the use of genetic variation in any other gene identified in the assessment of risk or in the diagnosis of Alzheimer's disease.
This work was supported by federal grants from the National Institute on Aging of the National Institutes of Health (P01AG07232, R37AG15473, P50 AG08702) and by grants from the Alzheimer Association, the Blanchette Hooker Rockefeller Fund, the Robertson Gift from the Banbury Fund and the Merrill Lynch Foundation.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.