Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Neurology. Author manuscript; available in PMC 2010 July 1.
Published in final edited form as:
PMCID: PMC2895568

Toward a plasma marker for Alzheimer disease: Some progress, but still a long way to go

Progress in neurology frequently lags 10 to 20 years behind that of cardiology. Early data correlating cardiovascular disease and serum levels of total and low-density lipoprotein cholesterol have led to treatments (exercise, weight loss, and statins) which have decreased age-adjusted cardiovascular mortality rates by 40%. In this regard, serum cholesterol levels have transitioned from a biomarker to a risk factor as treatment aimed at lowering serum levels have conferred clinical benefits. For 25 years, neurologists have been trying to apply this model to Alzheimer disease (AD) with little success. In the current issue of Neurology,® Ertekin-Taner et al.1 add to the growing literature that plasma Aβ42levels are a potential biomarker for AD.26 Because Aβ42 was elevated in plasma in asymptomatic first-degree relatives of patients with presenilin 1, presenilin2, and amyloid precursor protein mutations who had early onset AD (EOAD), they reasoned that Aβ42 levels might also be elevated in asymptomatic first-degree relatives of patients with late onset AD (LOAD).

The Jacksonville Mayo group compared plasma Aβ42 levels between two cohorts of relatives of patients with AD and two groups of controls. The first cohort consisted of 103 first-degree relatives of patients with LOAD and 106 unrelated controls, the second of 109 subjects from 25 multigenerational LOAD families and 50 married-in spouses. In both cohorts, the relatives were between 20 and 65 years of age and cognitively normal. In both cohorts, small but significant differences between relatives and controls were found. However, there are several important caveats. First, the absolute magnitude of the differences was small. Second, measurements of Aβ42 required substantial normalization, and comparisons of absolute values across different assay batches were not possible. Indeed, the absolute value in family members in one study was virtually the same as that of controls in a second study. Third, although the Mayo group showed only modest variation in Aβ42 levels on a time scale of hours or days, substantial increases occur with normal aging after the age of 65. In terms of predictive value in individual family members, the utility of Aβ42 remains to be demonstrated.

Among patients with dementia and clinical Alzheimer disease, plasma Aβ42 levels are lower than normal, suggesting that plasma levels gradually decline from elevated levels through normal levels to low level as patients transition from cognitively normal with increased genetic risk through MCI and into clinical dementia. Thus to be clinically useful, clinicians would need to do long term tracking of Aβ42 levels.

Their study also adds indirect evidence concerning the hereditability of LOAD. Genetic studies of EOAD where many cases are autosomal dominant quickly led to finding of mutations in the amyloid precursor protein, presenilin I, and presenilin II genes. These findings were strongest support for the amyloid cascade hypothesis as each mutation led to increased synthesis of Aβ42. However, EOAD—arbitrarily defined as onset before age 65—accounts for less than 5% of all cases of AD. What are the genetic contributions to LOAD?

In 1993, Corder et al.7 reported that apolipoprotein ε4 was associated with increased risk of AD and earlier age at onset. The apolipoprotein ε4 allele was not fully penetrant—subjects homozygous for the allele have lived into their 100s and never develop AD. Only 40% of all patients with LOAD have even one apolipoprotein ε4 allele, yet to date no other LOAD gene has been conclusively proven and some have doubted whether “another apoE” where a common polymorphism contributes to a significant share of LOAD cases will be found.8 The current work of Ertekin-Taner et al. demonstrates that the inheritability of elevated plasma Aβ was 31.9 ± 13.8% (p as shown < 0.007) using a conservative analysis model. This group had previously shown that genetic mutations that cause EOFAD increase plasma Aβ42, so the variants responsible for the hereditability of Aβ would be expected to contribute to the hereditability of LOAD, but the precise relationship between the inheritability of LOAD and plasma Aβ remains to be determined.

In contrast to biomarkers in blood, CSF biomarkers are already used by some in the diagnosis of AD and are certainly closer than blood markers to being of genuine clinical usefulness. CSF levels of hyperphosphorylated tau and Aβ1–42 have been predictive of conversion from MCI to AD.9 At first one might think that lumbar puncture is too invasive to gain widespread acceptance as a screening test in patients whose only risk for AD is advancing age. Yet lumbar puncture may be no more invasive than colonoscopy, which is the standard of care as a screening test for colon cancer in adults over age 50. Nonetheless, the utility of a marker in plasma or serum that can be obtained repeatedly is self-evident.

The report by Ertekin-Taner et al.1 adds yet another small set of data supporting the “amyloid cascade hypothesis.” Longitudinal studies testing whether cognitively normal first-degree relatives with elevated Aβ42 levels truly do develop AD more frequently than cognitively normal first-degree relatives with normal Aβ42 levels have not been completed. Until such studies are completed and replicated, plasma Aβ42 levels must remain one of many potential biomarkers. In the meantime, plasma Aβ42 has proven to be a surrogate quantitative phenotype for AD that is useful in identifying candidate genes associated with AD.


Disclosure: Peter Davies is a consultant to Applied Neurosolutions Inc., Vernon Hills, IL.


1. Ertekin-Taner N, Younkin LH, Yager DM, et al. Plasma amyloid β protein is elevated in late-onset Alzheimer disease families. Neurology. 2008;70:596–606. [PubMed]
2. Schupf N, Patel B, Pang D, et al. Elevated plasma β-amyloid peptide Aβ42 levels, incident dementia, and mortality in Down syndrome. Arch Neurol. 2007;64:1007–1013. [PMC free article] [PubMed]
3. Mayeux R, Honig LS, Tang MX, et al. Plasma Aβ40 and Aβ42 and Alzheimer's disease: relation to age, mortality, and risk. Neurology. 2004;61:1185–1190. [PubMed]
4. Pomara N, Willoughby LM, Sidtis JJ, et al. Selective reductions in plasma Aβ1–42 in healthy elderly subjects during longitudinal follow-up: a preliminary report. Am J Geriatr Psychiatry. 2005;13:914–917. [PubMed]
5. van Oijen M, Hofman A, Soares HD, et al. Plasma Aβ1–40 and Aβ1–42 and the risk of dementia: a prospective case-cohort study. Lancet Neurol. 2006;5:655–660. [PubMed]
6. Graff-Radford NR, Cook JE, Lucas J, et al. Association of low plasma Aβ42/Aβ40 ratios with increased imminent risk for mild cognitive impairment and Alzheimer's disease. Arch Neurol. 2007;64:354–362. [PubMed]
7. Corder EH, Saunders AM, Strittmatter WJ, et al. Gene dose of apolipoprotein E type 4 allele and the risk of Alzheimer's disease in late onset families. Science. 1993;261:921–923. [PubMed]
8. Tanzi RE, Bertram L. Twenty years of the Alzheimer's disease amyloid hypothesis: a genetic perspective. Cell. 2005;120:545–555. [PubMed]
9. de Leon MJ, Mosconi L, Blennow K, et al. Imaging and CSF studies in the preclinical diagnosis of Alzheimer's disease. Ann NY Acad Sci. 2007;1097:114–145. [PubMed]