Biomarkers for Alzheimer’s Disease: Are We Ready or Not?

Biomarkers have been developed in general medicine to diagnose, prognose and monitor disease progression. An ideal biomarker should detect a fundamental feature of the underlying pathophysiology of a disease and distinguish that illness from other conditions with an acceptable positive and negative predictive value.

Furthermore, the biomarkers should be reliable, relatively non-invasive, simple to perform and inexpensive. These criteria, outlined by a workgroup report about the use of biomarkers in Alzheimer’s disease (AD) in 1998 (ref. 1) are currently not met by any single marker, but they represent the high standards to which the research field aspires.

Over the last decade, however, Alzheimer’s disease has seen tremendous growth of candidate biomarkers.

Neuropsychological testing

In this report,we will briefly review the evidence for emerging biomarkers across multiple biological platforms, including neuropsychological testing, blood tests, genetic markers, cerebrospinal fluid and brain imaging.

To begin, neuropsychological testing has long been the mainstay diagnostic and disease severity marker for AD (ref. 2). Although neuropsychological tests are not generally considered as classic biomarkers, such tests certainly track the underlying condition and have often been used as dependent variables in drug efficacy trials (ref. 3).

In fact, changes in neuropsychological testing are an integral part of the diagnostic criteria for AD and Mild Cognitive Impairment (MCI), a possible prodromal version of AD (ref. 4 & ref. 5). Neuropsychological tests that are predictive of AD have been sought for decades without great success, though there are suggestions of some subtle changes that may precede the onset of clinical illness (ref. 6).

Many candidate tests have been proposed as early markers (ref. 7); still, no single test or battery of tests has fulfilled the criteria required of a standardized biomarker.

"Trait" variables and "state" variables

Genetic markers for AD have also been recognized for many years, and there are now multiple known mutations across three separate chromosomes with others likely to be identified in the future (ref. 8).

While these autosomal dominant mutations are fully penetrant, they do not in themselves identify the underlying biology of AD with any temporal accuracy, because the clinical phenotype is expressed only if the subjects live long enough to reach the age of vulnerability.

Strictly speaking, these specific mutations are therefore long-term prognostic biomarkers of AD and not diagnostic biomarkers. They are "trait" variables and not "state" variables. While not a mutation, APOE ε4 is a most interesting genetic marker that is located on chromosome 19. APOE ε4 has been found to be strongly associated with AD (ref. 9).

Presence of one or more ε4 alleles is associated with an increased risk of AD and is perhaps the most replicated finding in all of AD research. Unlike the genetic mutations, however, there is no absolute certainty of developing AD with increasing age, only an increased likelihood.

Nevertheless, APOE ε4 has become an important predictor of risk, and when combined with other markers such as neuropsychological testing, it has already proven to be of value in discerning potential subsets of individuals at even greater risk of developing AD (ref. 7).

Similarly, combining the risk associated with the APOE ε4 allele with neuroimaging measures has produced a potential for further refinement of identifying individuals at greater risk for developing AD (ref. 10; ref. 11).

Blood tests an attractive AD biomarker

Some blood tests available at baseline evaluation have also been proposed as markers of risk for AD. Serum homocysteine is an example of such a test, but the association with AD is relatively weak and needs further validation (ref. 12).

Vitamin B12 and folate are sometimes mentioned as biomarkers of AD, but they are more important as markers of other confounding illnesses and used primarily as markers of exclusion in the diagnostic evaluation of potential AD cases (ref. 13). Plasma β-amyloid is very attractive as a potential biomarker because of its natural association with the amyloid plaque and the known pathophysiology of AD (ref. 14)

However, there is much question concerning the variability of the plasma levels and whether they are significantly related to brain levels and the disease process itself; much more validation is needed before this measure can become an attractive AD biomarker.

Cerebrospinal fluid (CSF) is generally considered a more proximal measure of brain activity than plasma. Indeed, CSF levels of β-amyloid1-42, total tau and, most recently, phosphorylated tau protein have been proven reliably altered in AD versus controls (ref. 15; ref 16).

As before, much more work in this area is needed to establish the usefulness of such biomarkers as a predictive measure of disease development or progression. Nevertheless, there is already evidence that these CSF measures are as accurate (85 - 90%) in establishing a diagnosis of mild-to-moderate AD as the longitudinal clinical tracking of patients followed to autopsy (ref. 15, ref. 17, ref. 19).

Remarkable degree of accuracy

This degree of accuracy is remarkable given the fact that the CSF test is a cross-sectional, onetime measure available today; however, it should be noted that this high degree of accuracy is generally found only in comparison to normal controls.

The specificity of this biological marker is not yet as well established clinically when the comparison population is another group with non-AD dementia (i.e., fronto-temporal dementia, Lewy body dementia, vascular dementia and Parkinson’s disease).

There has also been some suggestion of CSF biomarker changes in MCI (ref. 20, ref. 21) and even in "at risk" controls (Sunderland et al., submitted) to help anticipate the onset of AD, but much more longitudinal data is needed to establish the link between these cerebrospinal fluid changes and the development of AD.

Surrogate markers of AD progression

Perhaps the best-studied category of biomarkers in AD and geriatric psychiatry in general is that of neuroimaging. Starting with rCBF scans and EEG studies in the 1980’s, researchers have been searching for imaging correlates of the brain deterioration in dementia, but there was only limited success.

More recently, structural imaging with magnetic resonance imaging (MRI) has been common, and there are numerous studies of various measures ranging from total brain volume to hippocampal volume that have been proposed as surrogate markers of AD progression (ref. 22; ref. 23).

While there are ongoing studies attempting to establish appropriate evidence for the specificity and sensitivity of these findings in AD populations, much more work is needed (ref. 24). For instance, it has not yet been established whether the rates of change are similar at different stages of the illness or whether there is a positive predictive value of early hippocampal changes seen in MCI subjects and people "at risk" for developing dementia.

Functional neuroimaging

Functional neuroimaging with positron emission tomography (PET) also has a relatively long history in AD research, dating back 20 years to measures of glucose metabolism (ref. 25). Functional MRI and PET studies have been conducted more recently, and they show signs of great promise, especially with concurrent cognitive testing (ref. 10; ref. 26; ref. 27).

Perhaps the most exciting development has been the addition of ligands to functional scanning, especially the markers that have the potential for measuring the brain β-amyloid1-42 or tangle burden (ref. 28; ref. 29). While other ligands have been examined previously, the β-amyloid1-42 ligand promises to be a tool to investigate pathophysiologicallyrelevant changes both during the course of illness and perhaps even before the clinical symptoms of AD are manifest.

Once more, the possibility of combining risk factors from various modalities (i.e., genetic, cerebrospinal fluid, neuropsychological testing and neuroimaging studies) offers a golden opportunity to maximize the individual risk factors and predict who might go on to develop the illness, but this work is still in its infancy.

Summary

In summary, the pool of potential biomarkers for the clinical diagnosis and therapeutic monitoring of AD is rapidly expanding, but it is not yet time for routine clinical application of these measures.

Larger-scale, longitudinal studies of AD subjects are needed to test the influence of disease severity and advanced pathology on the proposed markers. This kind of information is necessary before these biomarkers can become valid outcome variables in clinical trials.

As noted previously, it is also imperative to compare the biomarkers in AD patients against other related diseases to investigate whether or not the measures are diagnostic for AD or non-specific but sensitive markers of a general neurodegenerative process. In either case, the biomarkers could be valuable, but they would be more helpful if there is specificity for one illness or another.

Finally, the ultimate task is to learn whether these biomarkers have prognostic utility in people who are yet to develop clinical symptoms. Longitudinal follow-up studies of such "at risk" subjects will be required to establish such utility, and these studies are currently ongoing.

So, to answer the question we initially posed in the title about the readiness of biomarkers in AD, we are close but not quite ready for prime time utilization of biomarkers for AD diagnosis and therapy in a routine clinical setting. The research in this area is exciting and promising, but there are too many questions still unanswered.

References

1. Consensus report of the Working Group on: ''Molecular and Biochemical Markers of Alzheimer’s Disease''. The Ronald and Nancy Reagan Research Institute of the Alzheimer’s Association and the National Institute on Aging Working Group. Neurobiol Aging 1998; 19 (2); 109-16.

2. Mohs RC, Cohen L. Alzheimer's Disease Assessment Scale (ADAS). Psychopharmacol Bull 1988; 24 (4); 627-8.

3. Rogers SL, Farlow MR, Doody RS, Mohs R, Friedhoff LT. A 24-week, double-blind, placebocontrolled trial of donepezil in patients with Alzheimer’s disease. Donepezil Study Group. Neurology 1998; 50 (1); 136-45.

4. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. APA 1994.

5. Petersen RC, et al. Mild cognitive impairment: clinical characterization and outcome. Arch Neurol 1999; 56 (3); 303-8.

6. Snowdon DA. Healthy aging and dementia: findings from the Nun Study. Ann Intern Med 2003; 139 (5 pt. 2); 450-4. (Note: Free full text article)

7. Levy J, et al. Context-specific memory and apolipoprotein E (ApoE) E4: Cognitive evidence from the NIMH prospective study of risk for Alzheimer’s disease. J Int Neuropsych Soc 2004; 10 (3); 362-370.

8. Tanzi RE, Bertram L. New frontiers in Alzheimer’s disease genetics. Neuron 2001; 32 (2); 181-4.

9. Corder EH, et al. Gene dose of apolipoprotein E type 4 allele and the risk of Alzheimer’s disease in late onset families. Science 1993; 261 (5123); 921-3.

10. Small GW, et al. Cerebral metabolic and cognitive decline in persons at genetic risk for Alzheimer’s disease. PNAS 2000; 97 (11); 6037-6042. (Note: Free full text article)

11. Reiman EM, et al. Functional brain abnormalities in young adults at genetic risk for late-onset Alzheimer’s dementia. PNAS 2004; 101 (1); 284-9. (Note: Free full text article)

12. Seshadri S, et al. Plasma homocysteine as a risk factor for dementia and Alzheimer’s disease. N Engl J Med 2002; 346 (7); 476-83. (Note: Free full text article)

13. Wang HX, et al. Vitamin B(12) and folate in relation to the development of Alzheimer’s disease. Neurology 2001; 56 (9); 1188-94.

14. Hardy J, Selkoe DJ. The amyloid hypothesis of Alzheimer’s disease: progress and problems on the road to therapeutics. Science 2002; 297 (5580); 353-6.

15. Sunderland T, et al. Decreased beta-amyloid 1-42 and increased tau levels in cerebrospinal fluid of patients with Alzheimer disease. JAMA 2003; 289 (16); 2094-103.

16. Hampel H, Goernitz A, Buerger K. Advances in the development of biomarkers for Alzheimer’s disease: from CSF total tau and Abeta(1-42) proteins to phosphorylated tau protein. Brain Res Bull 2003; 61 (3); 243-53.

17. Mirra SS, et al. The Consortium to Establish a Registry for Alzheimer’s Disease (CERAD). Part II. Standardization of the neuropathologic assessment of Alzheimer’s disease. Neurology 1991; 41 (4); 479-86.

18. Galasko D, et al. High cerebrospinal fluid tau and low amyloid beta42 levels in the clinical diagnosis of Alzheimer disease and relation to apolipoprotein E genotype. Arch Neurol 1998; 55 (7); 937-45.

19. Blennow K. Cerebrospinal Fluid Protein Biomarkers for Alzheimer’s Disease. J Am Soc Exp Neuro Therap 2004; 1 (2); 213-225.

20. Andreasen N, Blennow K. Beta-amyloid (Abeta) protein in cerebrospinal fluid as a biomarker for Alzheimer’s disease. Peptides 2002; 23 (7); 1205-14.

21. Hampel H, et al. Value of CSF beta-amyloid (1-42) and tau as predictors of Alzheimer’s disease in patients with mild cognitive impairment. Mol Psychiatry 2004; 9 (7); 705-10.

22. de Leon MJ, et al. In vivo structural studies of the hippocampus in normal aging and in incipient Alzheimer’s disease. Ann NY Acad Sci 1996; 777; 1-13.

23. Cohen RM, Small C, Lalonde F, Friz J, Sunderland T. Effect of apolipoprotein E genotype on hippocampal volume loss in aging healthy women. Neurology 2001; 57 (12); 2223-8.

24. Mortimer JA, Gosche KM, Riley KP, Markesbery WR, Snowdon DA. Delayed recall, hippocampal volume and Alzheimer neuropathology: findings from the Nun Study. Neurology 2004; 62 (3); 428-32.

25. Cutler NR, et al. Brain metabolism as measured with positron emission tomography: serial assessment in a patient with familial Alzheimer’s disease. Neurology 1985; 35 (11); 1556-61.

26. Reiman EM, et al. Declining brain activity in cognitively normal apolipoprotein E epsilon heterozygotes: A foundation for using positron emission tomography to efficiently test treatments to prevent Alzheimer’s disease. PNAS 2001; 98 (6); 3334-3339.

27. Devous MD, Sr. Functional brain imaging in the dementias: role in early detection, differential diagnosis, and longitudinal studies. Eur J Nucl Med Mol Imaging 2002; 29 (12); 1685-96.

28. Small GW, et al. In vivo brain imaging of tangle burden in humans. J Mol Neurosci 2002; 19 (3); 323-7.

29. Klunk WE, et al. Imaging brain amyloid in Alzheimer’s disease with Pittsburgh Compound-B. Ann Neurol 2004; 55 (3); 306-19.

(This article was first published by the Lundbeck Institute in the Institute Magazine no. 8, 2004)