Biomarkers for Alzheimer’s disease, NRDD 2010

Basic and clinical research advances over the past decades have provided detailed knowledge of the molecular mechanisms and clinical course of Alzheimer’s disease. Alzheimer’s disease is a complex progressive condition that involves sequentially interacting pathological cascades, including the interaction of amyloid-β (Aβ) aggregation with plaque development, and the hyperphosphorylation and aggregation of tau protein with formation of tangles. Together with associated processes, such as inflammation and oxidative stress, these pathological cascades contribute to loss of synaptic integrity and progressive neurodegeneration1. These advances in research have been translated into several new drug candidates with diseasemodifying potential, many of which are now being evaluated in clinical trials2. Studies in transgenic mouse models of Alzheimer’s disease suggest that the majority of these new types of diseasemodifying drugs may be most effective early on in the process of Aβ aggregation, and be less effective in later stages when there is severe plaque pathology and neurodegeneration3–5. However, the current diagnostic criteria (DSM-IV, ICD-10 and NINCDS-ADRDA) that are used to identify patients with Alzheimer’s disease who

have overt dementia correspond to neuropathologically advanced disease. Consequently, early recognition of the disease needs to be improved; indeed, it is estimated that interventions that could delay the clinical onset of dementia by 1 year would reduce the prevalence in 2050 by 9 million cases6.

The primary end points in current clinical trials for Alzheimer’s disease measure the symptoms of the disease, such as cognitive and functional impairment. However, performance in these tests is determined by multiple factors other than key pathological events, which lead to neurodegeneration. As a result, drug trials that use clinical rating scales as outcome measures will fail to define the impact of new drugs on Alzheimer’s disease pathology.

There is a growing need for biomarkers of Alzheimer’s disease pathology to improve drug development related to the disorder. Animal models of Alzheimer’s disease have low predictive power for determining the efficacy of treatments in patients with sporadic Alzheimer’s disease1. Human biomarkers that facilitate the identification of the biochemical effects of a drug in shortterm pilot studies may identify those drug candidates derived from animal

*Department of Psychiatry, Psychosomatic Medicine and Psychotherapy, Johann Wolfgang Goethe-University, Heinrich-Hoffmann-Str. 10, 60528 Frankfurt/Main, Germany.Correspondence to H.H. e-mail: [email protected]:10.1038/nrd3115

Amyloid-β(Aβ). An aggregation-prone peptide derived from the amyloid precursor protein. The 42 amino acid isoform of the peptide is the main component of plaques in Alzeimer’s disease.

Biomarkers for Alzheimer’s disease: academic, industry and regulatory perspectivesHarald Hampel*, Richard Frank‡, Karl Broich§, Stefan J. Teipel||¶, Russell G. Katz#, John Hardy**, Karl Herholz‡‡, Arun L. W. Bokde§§, Frank Jessen¶||||, Yvonne C. Hoessler¶¶, Wendy R. Sanhai##, Henrik Zetterberg***, Janet Woodcock‡‡‡ and Kaj Blennow***

Abstract | Advances in therapeutic strategies for Alzheimer’s disease that lead to even small delays in onset and progression of the condition would significantly reduce the global burden of the disease. To effectively test compounds for Alzheimer’s disease and bring therapy to individuals as early as possible there is an urgent need for collaboration between academic institutions, industry and regulatory organizations for the establishment of standards and networks for the identification and qualification of biological marker candidates. Biomarkers are needed to monitor drug safety, to identify individuals who are most likely to respond to specific treatments, to stratify presymptomatic patients and to quantify the benefits of treatments. Biomarkers that achieve these characteristics should enable objective business decisions in portfolio management and facilitate regulatory approval of new therapies.

b i o m a r k e r s

R E V I E W S

560 | july 2010 | Volume 9 www.nature.com/reviews/drugdisc

© 20 Macmillan Publishers Limited. All rights reserved10

http://www.ncbi.nlm.nih.gov/omim/104300

mailto:[email protected]

mailto:[email protected]

Tau proteinA microtubule-associated protein located in the neuronal axons. Hyperphosphorylated tau is the main component of neurofibrillary tangles in Alzheimer’s disease.

DSM-IV(Diagnostic and Statistical Manual of Mental Disorders, Fourth edition). This manual outlines diagnostic criteria for psychiatric disorders and is published by the American Psychiatric Association.

ICD-10(International Statistical Classification of Diseases, Tenth edition). This book outlines diagnostic criteria for human diseases and is published by the World Health Organization.

models that affect the disease process in patients. In clinical academic research, when qualified for use (BOX 1) in a specific patient population, biomarkers should aid in population selection and assessment of drug effects on disease progression. In industryled drug discovery and development, biomarkers should facilitate selection of drug candidates, verify the mechanism of action, define dose effects and enable clinical trials to be shortened and run with reduced sample size. Biomarkers could serve as surrogate end points (discussed in more detail below) for clinical outcomes, and so could increase objectivity and efficiency in regulatory decisionmaking. These could include, for example, labelling decisions in which a drug could be labelled as ‘disease modifying’ rather than just providing symptomatic treatment7. In addition to this, biomarkers should serve as diagnostic tools in clinical practice to allow early and presymptomatic identification of patients with Alzheimer’s disease, and to aid treatment decisions and monitoring in individualized care. Finally, biomarkers could serve as screening tools for disease prevention programmes.

To qualify for these purposes, a biomarker must be measured by reliable and validated methods, must be sensitive and specific when used as diagnostic markers, must be sensitive to the effects of a drug, and must be predictive of clinical outcomes8,9.

In clinical trials, biomarkers that are indicators of the central or downstream elements of Alzheimer’s disease pathogenesis could serve at least three different purposes:

as diagnostic biomarkers, to detect and monitor effects of drug candidates on the disease process, and as safety markers to detect and monitor potential side effects of drug candidates at an early stage. The last item may be of particular importance in clinical trials of novel drugs against Alzheimer’s disease, given the risk of encephalitis associated with monoclonal antibodies10. Biomarkers can be compounds obtained from bodily fluids or tissue, such as cerebrospinal fluid (CSF) assays11, or technically derived correlates of Alzheimer’s disease pathology, such as brain imaging markers. In particular, neuroimaging techniques12 have been developed that provide evidence for Aβ deposition, tau aggregation and neurodegeneration already at very early clinical disease stages13. However, the current generation of biomarkers are not yet robust against standard criteria that biomarkers in other branches of medicine fulfil. Also, the reporting of studies involving diagnostic biomarkers in Alzheimer’s disease needs to follow strict quality criteria, such as the quality assessment of diagnostic accuracy studies (QuADAS) criteria14.

Here, we review the current status of multimodal core biomarker candidates derived from structural, functional and metabolic neuroimaging, and from neurochemistry and genetic studies of Alzheimer’s disease. We also provide the converging perspectives of industry stakeholders and regulatory agencies on biomarker discovery and development. In our opinion, the integration of scientific knowledge and united international and interdisciplinary research efforts by academic institutions, the pharmaceutical industry and the regulatory agencies will accelerate the discovery and codevelopment of new and more informative biomarkers for broad clinical diagnostic use, as well as for the many trials of innovative treatments in neurodegenerative diseases and Alzheimer’s disease.

imaging as an end point in clinical trialsFour imaging modalities have been used as secondary end points in clinical trials on Alzheimer’s disease: structural magnetic resonance imaging (mRI), functional mRI (fmRI), magnetic resonance spectroscopy (mRS) and positron emission tomography (PeT). In structural mRI studies, there are correlations between mRIbased volume and neuron numbers in specific brain regions15. The blood oxygendependent level (BolD) fmRI signal is primarily a measure of the input and processing of neuronal information within a brain region16. mRS represents changes in the biochemical composition of the brain tissue. PeT using18F2fluoro2deoxydglucose (FDG) is thought to represent neuronal glucose consumption as the main determinant of neuronal metabolism17. PeT using tracers for Aβ is thought to represent accumulation of the pathognomonic sign of Alzheimer’s disease, that is, Aβ plaques18.

The validity of a biomarker with respect to a supposed neurobiological substrate will be relevant for the evaluation of diseasemodifying treatments. The imaging techniques provide information on the regional distribution of changes on a macroscopic (fmRI, PeT, mRS) or a mesoscopic (mRI) scale. Such knowledge on the spatial distribution and temporal dynamic changes of the brain

author addresses‡Global Medical Affairs and Clinical Strategy, GE Healthcare, 101 Carnegie Center, Princeton, New Jersey 08540, USA.§Federal Institute for Drugs and Medical Devices (BfArM), Kurt-Georg-Kiesinger-Allee 3, D-53175 Bonn, Germany.||Department of Psychiatry, University of Rostock, Gehlsheimer Str. 20, 18147 Rostock, Germany.¶DZNE, German Center for Neurodegenerative Disorders, Gehlsheimer Str. 20, 18147 Rostock, Germany.#Division of Neurology Products, Center for Drug Evaluation and Research, Food and Drug Administration, 10903 New Hampshire Avenue, Silver Spring, Maryland 20993, USA.**Department of Molecular Neuroscience and Reta Lila Weston Laboratories, Institute of Neurology, University College, Queen Square House, Queen Square, London WC1 3BG, UK.‡‡Wolfson Molecular Imaging Centre, University of Manchester, 27 Palatine Road, Manchester M20 3LJ, UK.§§School of Medicine and Trinity College Institute of Neuroscience, Trinity College Dublin, Trinity Centre for Health Sciences, The Adelaide and Meath Hospital Incorporating The National Children’s Hospital, Tallaght, Dublin 24, Ireland.||||Department of Psychiatry, Rheinische Friedrich-Wilhelm University Bonn, Sigmund-Freud-Strasse 25, 53105 Bonn, Germany.¶¶Department of Psychiatry, Alzheimer Memorial Center, Ludwig-Maximilian University, Nussbaumstr. 7, 80336 Munich, Germany.##Senior Scientific Advisor, Office of the Commissioner, Food and Drug Administration, 10903 New Hampshire Avenue, Silver Spring, Maryland 20993, USA.***Department of Psychiatry and Neurochemistry, Institute of Neuroscience and Physiology, The Sahlgrenska Academy at University of Gothenburg, S-431 80 Mölndal, Sweden.‡‡‡Center for Drug Evaluation and Research, Food and Drug Administration, 10903 New Hampshire Avenue, Silver Spring, Maryland 20993, USA.

R E V I E W S

nATuRe ReVIeWS | Drug Discovery Volume 9 | july 2010 | 561


NINCDS-ADRDA(The National Institute of Neurological and Communicative Disorders and Stroke and the Alzheimer’s Disease and Related Disorders Association). This outlines criteria for diagnosing Alzheimer’s disease.

Primary end pointA primary end point is defined as the single main question to be answered in a given clinical trial.

BiomarkerAn objective measure of a biological or pathogenic process that can be used to evaluate disease risk or prognosis. It can be used to guide clinical diagnosis or to monitor therapeutic interventions.

Surrogate end pointA substitute for a clinical end point in a clinical trial.

Secondary end pointAn end point in a trial that provides additional characterization of treatment effect, but is not sufficient by itself to fully characterize the benefit or to support a claim for a treatment effect.

in Alzheimer’s disease is important. This is because a systematic brain disease that follows a specific pattern of progression through the brain may lead to differences in the extent of pathological changes depending on the stage of the disease. Additionally, imaging biomarkers have been used as predictive markers of the progression of dementia in defined risk groups of patients. For example, biomarkers can increase the accuracy of predicting the conversion of mild cognitive impairment — a clinically defined risk syndrome in which normal cognition and dementia cannot be sharply separated — to dementia19,20.

Structural MRI. In Alzheimer’s disease, structural mRI typically shows a pattern of decreased grey matter in the parahippocampal gyrus, the hippocampus, the amygdala, the posterior association cortex and the subcortical nuclei including the cholinergic basal forebrain. In longitu-dinal studies mRI can be used as a potential marker to discriminate between diseasemodifying and symptomatic treatment effects by determining the rate of atrophy of brain regions (FIG. 1). TABLE 1 and Supplementary information S1 (table) outline the characteristics of different mRIbased disease markers.

The reliability of volumetric measures obtained from repeated mRI scans is generally high21, which is an important prerequisite for its use as a disease progression marker. Additionally, multicentre variability of manual and automated volumetric measures across 12 different mRI scanners was below 5% in one study22. To reduce the variability of mRI measures between different scanners in clinical trials, a phantom test is required to ensure that participating centres meet a set of minimal criteria for scanner quality. The most commonly mRIderived measure is hippocampus volume, which is measured by visual inspection or manual drawing on mRI slices.

However, more recent approaches use automated spatial transformation of mRI volumes in a common standard space and derive univariate or multivariate statistics from maps of cortical grey matter density, cortical thickness, white matter density or highresolution spatial transfor-mation maps. These approaches are less labour intensive than manual volumetric analyses and so far have been applied as secondary outcomes in few clinical trials. However, the multicentre reliability of these techniques has yet to be determined.

Pharmacological fMRI. Studies of memory in patients with Alzheimer’s disease using fmRI discovered a pattern of altered activation in the medial temporal lobes and parietal lobes, which were consistent with structural mRI results. Based on initial singlecentre studies, specific effects of treatment on regional brain activation could be detected in Alzheimer’s disease23 (FIG. 2; TABLE 1; see Supplementary information S1 (table)). multicentre and longitudinal studies using fmRI in patients with Alzheimer’s disease have yet to be done, which will be a major step in the further development of fmRIbased biomarkers.

Proton-MRS. ProtonmRS (1HmRS) provides quantitative biochemical measures of compounds in brain tissue. The best established 1HmRS marker is the amino acid Nacetyl aspartate (nAA), which reflects the functional status of neuronal mitochondria24. A reduction of nAA levels independent of brain atrophy is a consistent finding in Alzheimer’s disease. Preliminary singlecentre studies have demonstrated that nAA levels are responsive to pharmacological treatment, and multicentre application of this technique has been demonstrated (TABLE 1; see Supplementary information S1 (table)). However, largescale applications in clinical trials are still pending. In addition to nAA, several other metabolites — such as cholinecontaining compounds, creatine and phosophocreatine, myoinisitol, and gluatmate and glutamine — are detectable with 1HmRS, but their potential as biomarker candidates in Alzheimer’s disease is controversial and requires refined investigations25.

FDG-PET. FDGPeT measures local glucose metabolism as a proxy for neuronal activity at a resting state without the need for cognitive activation. In Alzheimer’s disease, neuronal activity is impaired, as FDG uptake is reduced, predominantly in temporoparietal association areas including the precuneus and posterior cingulate cortex. These changes are closely related to cognitive impairment as demonstrated in crosssectional and longitudinal studies. The changes in FDG uptake can be measured objectively with smaller coefficients of variance than standard neuropsychological measures, thus increasing the power of studies26,27. Alterations in FDG uptake are usually attributed to pharmacodynamic drug effects but will also reflect disease progression, especially when measured after several months of followup. This technique has been used as a secondary outcome parameter in some clinical trials28–31.

Box 1 | Validation and qualification of biomarkers

To facilitate communication among stakeholders it is useful to distinguish between the words ‘validation’ and ‘qualification’. Validation is often used to mean the determination of the performance characteristics of an assay — for example sensitivity and specificity — in measuring a particular analyte. Validation also has a jargonized use in the context of regulatory approval to market kits for commercial use as clinical diagnostics. Qualification is often used to mean the establishment of the credibility of that assay in its application to questions relevant to drug development. Such questions include “does the drug hit the target, and if so to what extent?”, “does the drug alter the mechanism by which it is intended to act” or “does the drug change pathophysiology in a clinically relevant way?”

Qualification requires specific patient populations and a specific therapeutic intervention. For example, a validated assay may be qualified as an Alzheimer’s disease biomarker for intervention with amyloid-β (Aβ) production but not in non-Aβ mechanisms. So, it could be said that the assay which was validated for quantification of Aβ plaques (or Aβ oligomers or Aβ monomers) in the brain or in CSF is ‘qualified for use’ as a biomarker in Alzheimer’s disease for drugs that inhibit β-site amyloid precursor protein-cleaving enzyme 1 (BACE1). However, as well as the assay validation data, additional clinical data are required to support such qualification. A validated assay might not be qualified for use as a biomarker, hence the distinction between validation and qualification (for use).

The ultimate use of a biomarker is as a surrogate end point, which requires that the biomarker has been qualified to substitute for a clinical standard of truth such that the biomarker reasonably predicts the clinical outcome and therefore can serve as a surrogate.

R E V I E W S



http://www.nature.com/nrd/journal/v9/n7/suppinfo/nrd3115.html

http://www.nature.com/nrd/journal/v9/n7/suppinfo/nrd3115.html

Nature Reviews | Drug Discovery

Cog

nitiv

e fu

nctio

n

Time

a Neuropsychology Treatment withdrawal

Disease modifyingSymptomaticPlacebo

Cog

nitiv

e fu

nctio

n

Time

b Atrophy Treatment withdrawal

Disease modifyingSymptomaticPlacebo

Longitudinal studyA research study with repeated observations of the same patients over long periods of time.

Phantom testA plastic cylinder with standardized measures and density used for calibration of magnetic resonance imaging devices.

Transformation mapA spatially extended vector field that describes the spatial warps that are needed to align a three-dimensional image of the brain into a common standard space.

Classical plaqueA dense aggregation of amyloid-β protein with classical amyloid characteristics, which is surrounded by swollen neuritis and reactive glial cells.

Amyloid-PET. Several studies have shown that selective binding of a11Clabelled thioflavin analogue known as Pittsburgh compound B (11CPIB)32 to Aβ can be visualized in PeT scans of patients with Alzheimer’s disease (TABLE 1; see Supplementary information S1 (table)). These amyloidPeT scans have provided information regarding the Aβ plaque burden that is independent from structural changes in brain anatomy33.11CPIB binding sites in the brain are associated with Aβ sheets in classical plaques and diffuse plaques, as well as cerebrovascular amyloid angiopathy18,34. It does not bind to soluble and oligomeric amyloid. Increased cortical uptake of 11CPIB is seen in 2 out of 3 of patients with mild cognitive impairment with probable progression to Alzheimer’s disease35,36. However, a significant proportion of elderly subjects without mild cognitive impairment show increased 11CPIB uptake with as yet unknown prognostic implications33,35. Currently, Phase II trials have been initiated with amyloid tracers that can be labelled with 18F isotopes, which have a longer halflife and wider availability compared with 11CPIB37.

Cholinergic neurotransmission. Cholinergic projections from the basal forebrain to cortical areas degenerate in Alzheimer’s disease, and treatment with choline esterase inhibitors has modest but proven efficacy. ligands for

choline esterase, cholinergic receptors and transporters have been used in human studies (TABLE 1; see Supplementary information S1 (table)), but have not yet been qualified as biomarkers.

Imaging biomarkers: future challenges. Presently, clinical trials include imaging end points on a projectbyproject basis. The standardization of imaging protocols and analysis techniques is paramount for acceptance in multicentre trials. In addition, regulatory authorities require proof that imaging outcomes are of relevance to drug safety or efficacy. A biomarker would serve as a surrogate end point if it reflects clinically relevant outcomes such as cognitive decline or health economic outcomes such as institutionalization. For instance, hippocampal atrophy could serve as a surrogate end point if hippocampus atrophy reflects the clinically relevant outcome of memory loss. This association needs to be further established in future studies.

nevertheless, an imaging marker could be used as a primary prespecified outcome measure in a proofofconcept study if the imaging marker reflects an underlying disease process. Hippocampus atrophy as determined by mRI correlates with hippocampal neuronal loss as shown in numerous clinical pathological studies15,38. Thus, hippocampus atrophy could be used to assess the mechanism of action of a new compound that claims to reduce neurodegeneration in Alzheimer’s disease. Therefore, imaging markers may be used to assess the presumed mechanism of action of a new compound in proofofconcept studies. For example, in the AN1792 vaccination trial39, as well as in the AlZHemeD Phase III trial40, there were higher rates of brain and hippocampus atrophy in the treated subjects than in the untreated controls. However, treated subjects showed no evidence of more rapid cognitive decline compared with the placebo groups. These examples illustrate that imaging end points can lead to unexpected findings that help to critically evaluate our primary assumptions on the mode of action of new compounds.

Genetic analysis in biomarker researchThe use of all the biomarkers discussed in this paper — and indeed all biomarkers for any disease — are limited in part by variability between cases of disease and variability between controls. This variability increases the sample sizes that are needed to achieve statistical significance and is notably higher, for many reasons including genetic variability, in humans compared to animal studies. So, genetic analysis of individuals is beneficial in four broad ways.

First, some biomarkers (such as Aβ42) came directly from genetic analysis of kindreds with the condition. As other genetic risk factors for Alzheimer’s disease are identified, it will be appropriate to assess their protein products, and other proteins, in the same pathway. Recent studies using large whole genome scans for Alzheimer’s disease41,42 identified clusterin (CLU; also known as apolipoprotein j) and complement receptor 1 (CR1) as risk loci, suggesting that these proteins themselves may be useful biomarkers for Alzheimer’s disease

Figure 1 | Detection of disease-modifying treatment effects. a | The solid red line indicates the course of cognitive decline (assessed by cognitive testing) during disease-modifying or symptomatic treatment compared with the course of decline seen with placebo (green line). Only after withdrawal from drug treatment in the second half of the trial does the disease-modifying treatment effect (upper dashed red line) differentiate from the symptomatic treatment effect (lower dashed orange line). b | The upper solid blue line indicates the course of atrophy in the brain (as measured by magnetic resonance imaging) with a disease-modifying compound compared with a symptomatic treatment (lower pale blue line) and placebo (green line). This difference is maintained after withdrawal of the drug, but the withdrawal phase of the trial would not be required to differentiate the disease-modifying effect from the symptomatic treatment effect based on a volumetric outcome.

R E V I E W S



http://www.ncbi.nlm.nih.gov/gene/1191


Diffuse plaqueAn aggregation of amyloid-β protein that can only be detected using immunohistochemistry.

AN1792 vaccination trialThe first clinical trial on active amyloid-β immunotherapy in Alzheimer’s disease.

(although admittedly apolipoprotein e (APOE) has not proved useful in this regard). These results also suggest that the complement cascade, of which both Clu and CR1 are a part of, should be investigated.

Second, it is well known that those who are either amyloid precursor protein (APP) or presenilin (PSEN) mutation carriers, or who have Down’s syndrome or are APOE ε4 allele homozygotes, have an extremely

high risk of developing Alzheimer’s disease at relatively welldefined ages. These defined populations offer the possibility of following the presymptomatic changes in biomarker levels and therefore correlating their behaviour with clinical state. This approach has been particularly informative and useful in the analysis of brain imaging biomarkers43–45, but also offers potential in the assessment of blood and CSF biomakers46.

Table 1 | imaging biomarkers for alzheimer’s disease

Method Measures cross-sectional findings

Longitudinal findings Perspectives

Magnetic resonance imaging (MRI)

Visual rating of hippocampus

High discrimination between Alzheimer’s disease versus control groups104

Poor detection of atrophy over time104

• The best-established structural imaging markers are hippocampus and whole brain volume, which already have been applied in clinical trials

• Other imaging markers may carry more comprehensive regional information, but require multicenter assessment before application in clinical trials

Manual volumetry of hippocampus

High discrimination between Alzheimer’s disease versus control groups105,106; prediction of Alzheimer’s disease in MCI with 70–80% accuracy107

Atrophy rates of 3–7% per year in patients with Alzheimer’s disease compared with 0.9% in healthy controls

Automated whole brain volumetry

Main application for longitudinal evaluation

Atrophy rate of 2.5% per year in patients with Alzheimer’s disease compared with 0.4–0.9% in healthy controls108

Automated measures of regional cerebral atrophy patterns

Consistent pattern of cortical and subcortical atrophy

Spread of atrophy throughout the brain in Alzheimer’s disease; effect sizes for application in clinical trials are difficult to derive

Functional magnetic resonance imaging (fMRI)

Blood oxygendependent level signal imaging during activation

Reliability of fMRI data is high within subjects between imaging sessions109

Specific effects of cholinergic treatment on regional brain activation in Alzheimer’s disease110–112

• Promising secondary end point owing to wide availability

• Initial studies on multicenter fMRI are positive

Magnetic resonance spectroscopy (MRS)

Single voxel proton spectroscopy of NAA

NAA in the hippocampus is reduced in Alzheimer’s disease and at the MCI stage in those subjects who later decline to dementia113

Increase of NAA during cholinergic treatment in Alzheimer’s disease114

• MRS marker provides complementary information to structural MRI and fMRI

• Has been evaluated in a first multicenter study

Positron emission tomography (PET)

Glucose (FDG) consumption

Typical pattern of parieto-temporal cortical hypometabolism; low variability between multiple centers115

Effects of cholinergic treatment on cortical metabolism in Alzheimer’s disease117,118

• FDG-PET is a stable and valid multicenter marker, but use is limited due to costs and availability

• The potential use of amyloid imaging with 11C-PIB and cholinergic imaging as surrogate end points is not clear

• Feasibility of both FDG-PET and 11C-PIB in multicenter studies has been demonstrated

Amyloid (11C- PIB)

High sensitivity in detecting amyloid plaques and vascular amyloid in human brains in vivo116

Little further increase of 11C-PIB uptake during progression of Alzheimer’s disease119

Acetylcholine esterase

Average reduction of cerebral acetylcholine esterase activity is in the range of 30–40%

Degree of acetylcholine esterase inhibition correlated with clinical improvement of cognition in Alzheimer’s disease120

11C- PIB, 11C-labelled Pittsburgh compound B; FDG, 18F-2-fluoro-2-deoxy-d-glucose; MCI, mild cognitive impairment; NAA, N-acetyl aspartate.

R E V I E W S






Huntington’s protocolIndividuals who have family members with Huntington’s disease are offered genetic counselling, which includes the possibility of presymptomatic DNA testing. However, this is only offered after a series of careful counselling sessions to ensure that the individual understands the issues involved in knowing their genetic status.

TOMM40A gene encoding a mitochondrial protein that is located on chromosome 19 adjacent to the apolipoprotein E gene.

Risk chartA graph illustrating the risk of developing Alzheimer’s disease by age. This risk is modified by genetic status, especially by apolipoprotein E genotype status. Other genes will have a smaller effect on this chart (except in families with amyloid precursor protein and presenilin mutations).

Third, one of the problems of biomarker research is normal variability in the level of the biomarker between individuals. At least some of this variability is genetic and can be factored out through genetic analysis. For example, expression levels of microtubuleassociated protein tau (MAPT) in the brain47 and tau levels in the CSF48,49 have been suggested to be influenced by the MAPT haplotype. This phenomenon is true of many proteins for which blood (and CSF) levels are directly influenced by genetic variability50.

Fourth, the incorporation of data relating to an individual’s response to a drug and to how the drug is metabolized (which can vary between individuals) into biomarkers and in clinical studies will be helpful51.

Genetic tests and Alzheimer’s risk. Genetic tests are available for individuals who have family members with earlyonset familial Alzheimer’s disease, and are offered using the Huntington’s protocol. These genetic screens include testing for PSEN1, PSEN2, APP and APOE mutations; however, as APP or PSEN2 mutations are rare, screening for these mutations currently has to be organized on an ad hoc basis. Although genetic testing of APOE and predictive testing in typical lateonset disease had been discouraged52, the predictive value of APOE ε4 homozygosity is as high as for many other mendelian diseases. Indeed, it seems that attitudes to APOE genotyping are changing53, and it is likely that the predictive potential of genetic analysis will be considerably increased. This is in part due to a growing appreciation of the increase in predictive value given by genotyping the whole APOE locus (including the TOMM40 region of the APOE promoter), together with hits from the recent genomewide association studies41,42.

A likely outcome of these endeavours is that these predictive tests will classify individuals at risk of developing Alzheimer’s disease into three groups. The first group includes individuals who are APOE ε4 homozygotes at high risk of developing Alzheimer’s disease. This risk is modulated by other genes and the polymorphisms of the APOE promoter (about 3% of the population). The second group includes individuals who are APOE ε4 heterozygotes who are at modest risk of developing Alzheimer’s disease. For this group, the precise risk is modulated by other APOE alleles (ε2 being less risky than ε3) and by the polymorphisms of the APOE promoter, as well as by the other genomewide association study hits. The third group includes individuals who do not carry an APOE ε4 allele and are therefore at low risk of developing Alzheimer’s disease. The interpretation of these data will need the development of a sophisticated risk chart, and their explanation to lay people will present a challenge. Although still to be investigated, it is expected that the adjunct use of these genetic tests will help considerably in the identification of individuals with mild cognitive impairment who will later be diagnosed with Alzheimer’s disease54.

As our understanding of the genetic architecture of Alzheimer’s risk increases through the identification of relatively common risk variants such as CLU, C1R and PICALM (phosphatidylinositol binding clathrin assembly protein), and the later identification of individually rare highrisk variants, we will be able to asymptotically construct a predictive algorithm of Alzheimer’s disease risk. Although this will considerably improve on what can be currently achieved, it is unlikely that this algorithm will be amenable for screening large atrisk populations (as indicated in our proposed diagnostic flow model depicted in FIG. 3).

biochemical biomarkers for use in clinical trialsThere are two main types of biochemical (that is, other than genomic or imaging) biomarkers. The first are core biomarkers that mirror fundamental pathogenic events in Alzheimer’s disease; for example, the deregulated metabolism of APP and Aβ. The second are downstream biomarkers that reflect secondary phenomena, such as axonal degeneration. Compared with CSF biomarkers, efforts to discover reliable biomarkers for Alzheimer’s disease in peripheral blood have been of limited success. Indeed, although there are many publications on potential candidate blood biomarkers, followup studies by other research groups are either lacking, or have failed to confirm a solid diagnostic value. nevertheless, a recent pilot study reported diagnostically useful patterns of 18 different signalling, acute phase and inflammatory proteins in plasma that warrant further investigation55. Plasma Aβ40 and Aβ42 can be measured in peripheral blood, but do not contribute to the identification of Alzheimer’s disease in a reproducible manner56, and are unlikely to reflect Aβ processing in the brain57. Instead, most research has focused on CSF biomarkers, which more directly reflects brain neurochemistry, and can be obtained through lumbar puncture without significant side effects58.

Figure 2 | Detection of effects of cholinergic treatment on cortical activation in patients with Alzheimer’s disease using functional magnetic resonance imaging. Regions of decreased activation (shown in blue) during a visual perception task in a group of patients with Alzheimer’s disease after 3 months of open-label treatment with an acetylcholinesterase inhibitor compared with activation before treatment. The visual perception task recruits parietal areas in healthy controls and the decrease in activation after treatment (blue areas) occurred in the cognitively relevant brain regions. Left side of image is left brain.

R E V I E W S







• Outcomes• Surrogate

• Monitor• Individualize

• Confirm• Characterize

• Enrich• Diagnose

New targetor indication

No

File dossiers

Yes

No

Yes


Newcandidates

Hits target

Affectspathophysiology

Clinicaldiagnosis

Treatment

No

Yes

Enhancedexposure

Alters mechanism

Or

Established CSF biomarkers. Four CSF biomarkers have been evaluated in a large number of independent studies: Aβ40, Aβ42, total tau and phosphorylated tau59. Aβ40, Aβ42 and phosphorylated tau reflect the core elements of the disease process in Alzheimer’s disease; that is, levels of brain amyloid60,61 and cortical neurofibrillary pathology62. Conversely, total tau is a nonspecific marker for axonal damage that mirrors the activity

of the neurodegenerative process11. The combination of elevated levels of total tau and phosphorylated tau together with reduced levels of Aβ42 or reduced Aβ42/Aβ40 ratio in CSF is a consistent finding in biomarker studies of patients with different stages of Alzheimer’s disease, including mild cognitive impairment11 (TABLE 2). A high diagnostic performance of these CSF biomarkers has been verified in three large multicentre studies, including the Alzheimer’s disease neuroimaging Initiative (ADnI) study63, the DeSCRIPA study64, and the Swedish Brain Power project65.

Potential CSF biomarkers. There are numerous other candidate biomarkers that reflect either elements of the primary pathogenic process in Alzheimer’s disease or secondary events of the disease. Biomarkers that mirror the pathogenic process include, for example, Aβ oligomers, βsite amyloid precursor proteincleaving enzyme 1 (BACe1) activity and concentration, secreted isoforms of APP, and Aβ degradation products66. Secondary events include oxidative stress responses, inflammation and gliosis67. The diagnostic potential of these biomarkers is less well studied. However, some biomarkers such as BACe1 activity68 and APP isoforms69 may also give important information on desired biochemical effects of certain drug candidates, such as BACe1 inhibitors.

Monitoring the biochemical effect of new drugs. Because there is slow and variable progression of symptoms in Alzheimer’s disease, very large cohorts of patients and a treatment duration of several years will be needed to identify a change in the rate of cognitive decline in clinical trials of diseasemodifying drug candidates. Small, shortterm trials that provide biochemical evidence of an effect of the drug on the central pathogenic pro cesses would therefore be of great value to make go/nogo decision before embarking on Phase II or III clinical trials.

Clinical treatment trials with acetylcholine esterase (AChe) inhibitors may serve as a proof of concept for the potential of CSF biomarkers to identify and monitor the biochemical effect of a drug in Alzheimer’s disease. Such studies found a marked increase in CSF AChe activity during treatment, which was dose dependent and linked to the mechanism of action of the drugs and the clinical outcome70. The mechanism for the increase in CSF AChe is unknown, but may involve an increased gene expression and release of AChe to the CSF, which is mediated by a muscarinic acetylcholine receptor feedback loop71.

The intraindividual variability of CSF tau proteins and Aβ42 is very low over time, with coefficients of variation of 4–6% during 6 months and 7–9% during 24 months of followup72,73. These results indicate that even minor changes in biomarker levels can be monitored. Studies in transgenic mice and in guinea pigs show reduced cortical, CSF and plasma Aβ levels during γsecretase inhibitor treatment74,75. The decrease in CSF Aβ also correlated with the decrease in cortical Aβ75. In a Phase II trial of the γsecretase inhibitor ly450139, acute treatment resulted in a dosedependent decrease in plasma Aβ levels, whereas after chronic treatment there was only a tendency

Figure 3 | The four categories of biomarker: target, mechanism, pathophysio-logical and diagnostic. Biomarkers can be categorized into four groups on the basis of their contribution to business, regulatory and clinical decision-making. Clinical decision-making can be further divided into clinical research and patient care diagnostic subcategories. The objective is to use biomarkers as early as possible in the drug development process. The initial step is to confirm that a test compound hits the target and to quantify the extent to which it does so. Next is to test three concepts in logical sequence. First, that hitting this target alters the pathophysiological mechanism. Second, that altering this mechanism affects the pathophysiology. Third, that affecting pathophysiology predictably improves the clinical status of the patients. Biomarkers qualified to confirm the presence of the target and or extent to which the drug candidate hits the target may be validated later as diagnostic tests for early detection or diagnosis of Alzheimer’s disease (when that target is expressed differentially between healthy and diseased states). Biomarkers qualified for confirming and quantifying mechanistic effects may be validated later as diagnostic tests to inform choice of therapeutic regimen, either in choice of drug or initial dosing regimen. Biomarkers qualified for longitudinal quantification of patient response in terms of clinically relevant pathophysiology, may be validated later as diagnostic tests for monitoring and individualization of a therapeutic regimen. Biomarkers qualified for either monitoring or individualization of therapy on clinically relevant pathophysiology may also serve as surrogate end points to support regulatory decision-making. In addition, they can be used to ensure appropriateness of use, and as quantifiers of clinical outcomes to support reimbursement decisions.

R E V I E W S



http://www.uniprot.org/uniprot/Q2HJ40

Table 2 | Core CsF candidate biomarkers for alzheimer’s disease

Analyte (method)

Analyte and assay characteristics

intra-individual variation

change in Alzheimer’s disease

evaluation notes

comments

APP/Aβ metabolism

Aβ42

(ELISA; Luminex; Meso Scale Discovery)

Confounding factors evaluated in several publications57,121; assay characteristics, including within-day and between-day CVs, well known57,121,122

Longitudinal CV for 6 months of 5%72 and CV for 24 months of 7%73

Reduction in levels by 50% in Alzheimer’s disease and in Alzheimer’s disease with MCI59

Consistent results, for reviews see REFS 11,66

CSF Aβ42

is the central CSF biomarker for Aβ metabolism

Aβ42

/Aβ40

ratio (ELISA)

Confounding factors evaluated in several publications66; assay characteristics, including within-day and between-day CVs, well known123–127

Stable Aβ40

levels during 12 months128 and 24 months129 of assessment

Reduction in ratio in Alzheimer’s disease is more marked than for Aβ

42 alone

Data based on a limited number of studies, for review see REF. 66

CSF Aβ42

/Aβ40

ratio may give a more accurate measure of amyloidogenic Aβ metabolism than Aβ

42 alone

APP isoforms: sAPPα, sAPPβ (ELISA; Meso Scale Discovery)

Assay characteristics and CVs published69,130

Not known No change found in Alzheimer’s disease69,130

Data based on a limited number of studies

APP isoforms are not diagnostically useful, but may be valuable in clinical trials on, for example, BACE1 inhibitors

BACE1 (enzyme activity assay)

Assay characteristics, including within-day and between-day CVs, known69,131

Not known Increase in BACE1 protein levels and activity in Alzheimer’s disease and in MCI68,69,132,133

Data based on publications using different methods

The diagnostic value of BACE1 activity has to be further evaluated, but may be useful in clinical trials on, for example, BACE1 inhibitors

Aβ oligomers (Bio-Barcode assay with PCR amplification134)

Not known Not known Increased CSF levels found in one pilot study134

Early method development

Aβ oligomers are highly promising biomarker candidates; however, very low CSF concentrations make method development difficult

Total Aβ turnover*

Not known Not known Not known Method publication based on healthy volunteers

Method may be valuable to gauge Aβ production and clearance in clinical trials; method development necessary to measure specific Aβ isoforms, that is, Aβ

42

and Aβ40

turnover

Total tau (T-tau)

T-tau (ELISA; Luminex)

Confounding factors evaluated in several publications122,135,136; assay characteristics, including within-day and between-day CVs, well known57,122,135


Marked increase in levels in Alzheimer’s disease and in Alzheimer’s disease with MCI11,59

Consistent results from numerous publications, for reviews see REFS 11,59

CSF T-tau is the central CSF biomarker to monitor cortical axonal degeneration in clinical treatment trials

Phosphorylated tau (P-tau)

P-tau 181 (ELISA; Luminex)

Confounding factors evaluated in several publications122,135,136; assay characteristics, including within-day and between-day CVs, well known57,122,135


Marked increase in levels in Alzheimer’s disease and in Alzheimer’s disease with MCI11,59

Consistent results from numerous publications, for reviews see REFS 11,59

CSF P-tau is the central CSF biomarker to monitor tau phosporylation state in clinical treatment trials

P-tau 231 (ELISA) Confounding factors and assay characteristics published137

Increased P-tau 231 levels during 12 months128 and 24 months129 of assessment

Marked increase in levels in Alzheimer’s disease and in Alzheimer’s disease with MCI11,137,138

Consistent results from numerous publications, for review see REF. 11

CSF P-tau is the central CSF biomarker to monitor tau phosporylation state in clinical treatment trials

Aβ, amyloid-β; APP, amyloid precursor protein; BACE1, β-site amyloid precursor protein-cleaving enzyme 1; CSF, cerebrospinal fluid; CV, coefficients of variation; MCI, mild cognitive impairment; sAPP, soluble APP. *Infusion of isotope-labelled leucine combined with continuous CSF sampling and immunoprecipitation, tryptic digestion and mass spectrometry measurement of Aβ.

R E V I E W S



Phase 0 trialAn exploratory first-in-human trial with single subtherapeutic drug doses and small numbers of subjects to provide first data on drug pharmacokinetics and pharmacodynamics.

for a reduction in CSF Aβ42 (REF. 76). Treatment of monkeys with a BACe1 inhibitor resulted in lower CSF Aβ42, Aβ40 and soluble APPβ levels77. Furthermore, a clinical trial on PBT2, which is suggested to affect metalinduced Aβ aggregation78, showed a dosedependent reduction in CSF Aβ42

levels79. The mechanism underlying the reduction in Aβ42 during treatment is unclear. Data from a small clinical study on the amyloidtargeting drug phenserine also suggest that CSF Aβ may be of value for evaluating treatment effects31.

Based on longitudinal studies of conditions that involve acute neuronal injury80,81 and data from the interrupted Phase IIa An1792 trial82, total tau should decrease towards normal levels if a treatment is successful in inhibiting the neurodegenerative process. The same may be expected for phosphorylated tau 181, which is supported by one pilot study of memantine83.

It is currently unknown whether changes in levels of biomarkers, especially for CSF Aβ42, during treatment will be different for Alzheimer’s disease cases who have low levels of CSF Aβ42 due to deposition in plaques at beginning of treatment. The degree and direction of change may also depend on the timepoint of sampling after initiation of treatment. In summary, these findings support the proposal that CSF biomarkers could be valuable tools to monitor biochemical drug effects in clinical trials.

industry perspective on biomarkersMonitoring the biochemical effect of new drugs. Biomarkers that are qualified for use in clinical trials to facilitate business and regulatory decisionmaking should also be available as diagnostic agents to enable appropriate prescribing. Therefore, each biomarker will have utility at one stage or another stage of medical product development; that is, from discovery to adoption in clinical practice (FIG. 4). moreover, the biomarkers could be used in our proposed diagnostic flow that incorporates risk assessment screening, diagnosis and prognosis, and monitoring treatment effect (FIG. 3). This diagnostic flow begins with tests that have high sensitivity but low specificity (and low cost) to those that have increasing specificity and potential for longitudinal quantification of treatment benefit.

What an Alzheimer’s disease biomarker could achieve in drug development. The determination of complete pharmacokinetic and pharmacodynamic relationships in relevant preclinical models shortens and aids objectivity in the selection of drug candidates and doses for clinical development. Given the costs associated with clinical development, the long duration of clinical trials relative to patent life and the difficulty in recruiting participants for clinical trials — which is partly due to the multiplicity of drugs being developed and other factors such as patient awareness — no drug candidate should enter the clinic without a speciesindependent biomarker as the central element of translational medicine (FIG. 4).

In addition to measuring pharmacological and toxicological effects during clinical trials, as well as efficacy and safety issues that arise postapproval, biomarkers facilitate stratification of the patients and dose optimization based on phenotype or genotype. moreover, enrichment

strategies for recruitment to the trial (that is, choosing only those most likely to gain benefit or least likely to suffer an adverse event) can shorten trials and improve the response rate. A successful example of this approach is HeR2/neu (also known as eRBB2) testing in breast cancer for trastuzumab (Herceptin; Genentech) treatment84.

Furthermore, biomarkers can be used for withinpatient dose titration during clinical trials, thereby effectively individualizing the therapeutic index. For example, a patient without amyloid plaques in the brain is less likely to respond to a drug that targets Aβ, whereas patients screened into the study with a positive plaque biomarker at baseline may not respond to the low starting dose, therefore warranting escalation to the next dose.

Identifying biomarkers at Phase 0. Currently, the efficacy of drugs for Alzheimer’s disease is assessed in patients clinically diagnosed with symptomatic Alzheimer’s disease. Although refinements in neuropsychological testing are facilitating earlier (presumptive) diagnoses, it is hoped that in the future Alzheimer’s disease can be diagnosed definitively before symptoms become apparent. It is at this stage that the drug has a better chance of efficacy, before the pathological changes in the brain are too advanced for therapeutic intervention. This is particularly important because of the long duration of pathophysiology changes before Alzheimer’s disease manifests in cognitive loss, and the shortcomings in clinical diagnosis85. For this purpose, biomarkers of early stages of Alzheimer’s disease are needed86.

There is currently no regulatory framework for which to approve drugs that could treat Alzheimer’s disease in its presymptomatic stages, so in anticipation of efficacy trials eventually being performed in earlystage disease, and in hope of drugs being approved for asymptomatic Alzheimer’s disease, Phase 0 trials of potential biomarkers should be started immediately so that the data are available to inform design of the Phase I–III trials in early disease87,88. Such a biomarker of earlystage pathophysiolgy would also enable treatment effects to be distinguished from symptomatic improvements, so that the drug can be labelled and prescribed appropriately (and priced accordingly). Such labelling could include disease modification claims that were based on the drug’s modification of the pathophysiological mechanism89. Conversely, if it is demonstrated that a drug inhibits the target without affecting downstream biomarkers, the drug class and target need reconsideration (TABLE 3). molecular imaging biomarkers are particularly valuable because they provide noninvasive anatomical specificity for Aβ plaques, tau and neurofibrillary tangles in animals and in humans.

Clinical use of Alzheimer’s disease biomarkers. The most valuable role for biomarkers has been identified by the uS Food and Drug Administration (FDA), and the national Cancer Institute90 as a clinical diagnostic (rather than a surrogate end point). A convergence is occurring between the requirements of biomarkers for quantification of drug effects in research and development, which are analysed as population means with

R E V I E W S




Post-treatment

Treatment-specificprognosis

Alzheimer’s disease-specific diagnosis

Early detectionSensitive, not specific

Screen for riskSensitive, not specific

Quantifying treatment effect enables individualization of regimen

Early, individualizedtreatment with disease-modifying drugs at optimaltherapeutic index

Diagnostics and clinicaldecision support informlikelihood of responseto treatment

Differential diagnosticsconfirm Alzheimer’s disease-specific pathophysiology

Diagnostic screening planstailored to risk profilechannel patients to morespecifc diagnostics

Genomics and family history inform screening plans and primary prevention strategies

Nonspecific, early symptoms

Asymptomatic

Asymptomatic

standard errors, and the requirements of diagnostics in clinical practice, which are assessed on a perpatient basis. The common element in this convergence is longitudinal quantification; both analyses require pretreatment and posttreatment effects of the drug to be measured. As diagnostics, biomarkers are of interest to payers and purchasers of health care for parallel applications. As an example, clinical evidence from the national oncologic PeT Registry resulted in expanded coverage by medicare for FDGPeT/CT. In addition to diagnosis and staging, a new ‘subsequent’ category was covered, which includes the monitoring of therapeutic benefit. Furthermore, earlier detection of disease facilitates earlier intervention, which, when followed by effective, individualized treatment, will yield better patient outcomes and reduced institutionalization. This is not unique to Alzheimer’s disease, but is particularly important because of the development of significant pathophysiology before the appearance of symptoms in Alzheimer’s disease, the inaccessibility of the lesions in the brain, and the need to distinguish between causes of dementia and then discern symptomatic improvement from disease modification. once a biomarker is proved to be ‘qualified for use’ (a phrase to be defined by regulatory guidance), withinpatient dose titration

can be incorporated into design of clinical trials and the resultant biomarker data could provide the information necessary for prescribing information to optimize outcomes for individuals by dose adjustments in clinical practice (TABLE 3). The biomarkers that are most likely to be utilized for the individualization of therapy are those that are useful in the longitudinal quantification of drug effects during research and development.

Who should pay for biomarker development? The costs and risks involved in qualifying biomarkers to satisfy the regulatory authorities, whether as surrogate end points or as a standard biomarker, may be prohibitive for a single company to invest in, but the alternatives of collaboration or public–private partnerships are complex. These costs, risks and complexities could be addressed by the issuance by the FDA, the european medicines Agency (emA) and other regulatory bodies, of harmonized guidelines for biomarker qualification that were more practical than unattainably pure statistical criteria that require capture of both efficacy and safety effects. Such guidelines would clearly outline the linkage between the degree of qualification required and the utility of the biomarker (that is, the biomarker would be qualified for use).

Figure 4 | Translation of biomarkers. The translation of biomarkers from being research tools that can be used in clinical trials, to being aids in regulatory decision-making and to being commercial diagnostics that can aid clinical decision-making facilitates the efficient selection of different patient populations. When used as research tools, biomarkers can identify a large population of individuals (that is, the base of the pyramid) who may benefit from low-cost and safe primary prevention strategies. Diagnostic biomarkers can be used in smaller populations with early forms of the disease which, when proven by a definitive diagnosis, warrants a more expensive and riskier therapeutic intervention that is tailored to their individual pathophysiology and monitored for efficacy (that is, the pinnacle of the pyramid). This role of biomarkers for diagnostics in individualization of therapy is also enabled by information technology, known as ‘clinical decision support’. Different biomarkers such as genomics and molecular imaging will each be useful in a diagnostic flow. This begins with risk assessment (enabling primary prevention) to screening (for early detection and early intervention), to diagnosis and prognosis (for staging and best choice of therapy) and finally to monitoring treatment effect (for true individualization of treatment). This flow starts from tests with high sensitivity but low specificity and low cost to those with increasing specificity and value with potential for longitudinal quantification. Patients will enter at different points in this diagnostic flow (from the bottom to the top of the pyramid) according to the manner in which they present; the two main options being with or without symptoms.

R E V I E W S



What creates disincentives? The qualification of a biomarker for disease modification requires that the biomarker is shown to capture the treatment benefit of a drug in development, that is, the drug candidate is a positive control. once this correlation between treatment benefit and biomarker has been achieved for two drugs with the same mechanism of action it might be construed as a surrogate end point for drugs with that mechanism of action. A biomarker that correlated with clinical benefit across drugs with differing mechanisms of action might be construed as a surrogate end point irrespective of mechanism of action.

Although companies may collaborate on Phase 0 trials, such as the ADnI (see Further information), trials that aim to measure treatment effect are inherently linked to specific compounds. The first company to succeed may enter the market earliest but will have taken all the risks, thereby effectively underwriting the qualification of that biomarker for competitors to use in reducing development costs, reducing regulatory risk, and increasing speed to market. When a second company successfully deploys that same biomarker, it might elevate it to surrogate end point status, but with the benefits again accruing to competitors and, ultimately, to the regulators and to patients.

An exclusivity provision analogous to that developed for paediatric trials (Guidance for Industry Qualifying for Pediatric Exclusivity Under Section 505A of the Federal Food, Drug, and Cosmetic Act) might compensate for these disincentives; however, more immediately, regulatory guidance on the qualification of biomarkers would be helpful, perhaps rendered less complex by making it specific to particular research platforms (for example, imaging as distinct from in vitro tests). meanwhile, the

qualification of a biomarker as a surrogate end point might be achieved after marketing. Guidance on codevelopment of therapeutics and diagnostics would also be helpful.

regulatory perspective on biomarkersFor regulatory purposes, the efficacy and safety of treatments are typically determined in at least two separate randomized, doubleblind, placebocontrolled trials that are each of at least 6 months duration. As a further requirement, efficacy must be established in a cognitive domain and a functional (for example, activities of daily living) or global domain (for example, clinical global impression of improvement). Significant differences must be demonstrated in both end points. These coprimary outcome measures are required for symptomatic and diseasemodifying approaches in pivotal Phase III studies. However, because of the pattern of symptom progression in Alzheimer’s disease and depending on the proposed mechanism of action of the drug, attempts to establish that a drug leads to a diseasemodifying effect are likely to require a much longer study duration and alternative trial designs91 compared with trials that measure symptomatic improvements.

As a consequence, the use of biomarkers for diseasemodifying approaches and the search for adequate surrogate end points is encouraged by all stakeholders involved in drug development in europe (for example, the Qualification of novel methodologies for drug development — Guidance to applicants by the emA, and the Improved Predictivity of Efficacy Evaluation — Brain Disorders by the Innovative medicines Initiative) and the united States92 (for example, the Critical Path Initiative). However, it is necessary to consider specific regulatory requirements when biomarkers are used as outcome measures to assess therapeutic agents — particularly those that lead to disease modification — for Alzheimer’s disease.

Biomarkers as surrogate end points. Although the use of biomarkers in earlier phases of drug development is well established, their value in pivotal efficacy studies for Alzheimer’s disease is still limited. A biomarker is a surrogate end point when it can be considered as a substitution for a clinically relevant end point that is a direct measure of how a patient feels, functions or survives, and can be expected to predict the effect of and allow measurement of the specific therapy93,94 (BOX 2) . However, even a perfect correlation between the level of marker with stage of disease in the untreated state is not sufficient to accept a biomarker as a surrogate for a clinical end point95,96. none of the imaging or biochemical markers has been sufficiently qualified as a surrogate end point in Alzheimer’s disease. There should be a link between a treatmentinduced change in the biomarker and the desired clinical outcome measure, as well as a link between the treatmentinduced change in the biomarker and change of disease process95,96. Additionally, based on the assumed mechanism of action of a given compound, there should be high plausibility (for example, based on preclinical models) that the disease process will be modified.

Table 3 | recommended biomarkers in clinical research and development

Biomarker entry criteria (early diagnosis)*

stratification (prognosis)*

Monitoring effect (individualization of treatment)*

APOE genotype – + –

Aβ42

, Aβ40

(plasma)‡ – – –/+

Aβ42

, Aβ40

(CSF) ++ + ++

Tau (CSF) ++ + ++

P-tau (CSF) ++ + ++

BACE1 (CSF) + –/+ –/+

MR volumetrics ++ – +

MR functional –/+ – –/+

MR spectroscopy – – –/+

FDG-PET ++ –/+ –

Amyloid-PET ++ ++ ++

– not useful; –/+ useful in limited circumstances, for example plasma amyloid-β (Aβ) for passive immunization; + generally useful; ++ always useful, as discerned from recent meetings of the Alzheimer’s Association Research Roundtable. APOE, apolipoprotein E; BACE1, β-site amyloid precursor protein-cleaving enzyme 1; CSF, cerebrospinal fluid; FDG, 18F-2-fluoro-2-deoxy-d-glucose; MR, magnetic resonance; P-tau, phosphorylated tau; PET, positron emission tomography. *Text in parentheses indicates potential application in clinical practice. ‡Measurements in plasma may be confounded due to other proteins present in plasma, and so may not accurately reflect the pathology.

R E V I E W S



http://www.fda.gov/downloads/Drugs/DevelopmentApprovalProcess/DevelopmentResources/UCM049924.pdf



http://www.ema.europa.eu/pdfs/human/biomarkers/7289408en.pdf


http://www.imi-europe.org/Lists/IMIPublicationDocuments/3_ Improved Predictivity of Efficacy Evaluation - Brain Disorders.pdf


http://www.fda.gov/ScienceResearch/SpecialTopics/CriticalPathInitiative/ucm204289.htm


http://www.alz.org/professionals_and_researchers_research_roundtable.asp

So, although the use of biomarkers as primary outcome measures (that is, surrogate end points) in pivotal efficacy trials is unlikely at this time, they can currently be used to allow decisionmaking on other aspects of drug development. For instance, they can be used in Phase II studies (for example, proof of concept, dose finding) as primary prespecified outcome measures or to better define patient populations at risk (for example, enriching populations likely to respond to therapy) for efficacy trials (BOX 3). The use of biomarkers in these ways may allow Phase II studies to be shorter or smaller in size, and if potential biomarkers can be qualified as surrogate end points, definitive effectiveness trials may also be considerably shorter and/or smaller than studies that use traditional clinical outcomes as primary end points.

Problems and concerns with using biomarkers as surro-gate end points. As noted above, the reliance on the drug’s pharmacological effect to modulate levels of a biomarker that has not been qualified for use as a surrogate end point has interpretative problems. For example, it is assumed that a treatment that is efficacious in Alzheimer’s disease will slow the progression of medial temporal lobe atrophy as measured by mRI. However, in the trial with the vaccine An1792 the extent of brain atrophy seemed to increase in patients that had increased antibodies to Aβ and that showed clinical improvement97. This was an entirely unexpected outcome, and provides evidence that the effects of a treatment (even a potentially beneficial treatment) on a biomarker can be unpredictable.

of more concern though are potential cases in which the desired effect on a nonvalidated biomarker is achieved. If the clinical effects of a medicinal product are unknown, then concluding that the drug has a beneficial effect for the patients would be based on the assumption that the desired effect seen on the biomarker will translate into the desired beneficial clinical effect. If this assumption is wrong, a drug that has no beneficial effect (or even, perhaps, a deleterious effect) might be approved. It is for this reason that approval of a treatment for Alzheimer’s disease based on an effect on a biomarker alone (that is, a surrogate end point) is unlikely at this time. However, as outlined in other sections, there is progress with some biomarkers towards validation for this purpose. As a general principle, if it is possible to ascertain effects of a drug on clinical outcomes in trials of reasonable size and duration, it is likely that approval of the drug will not be granted if the study has examined biomarkers as primary outcomes. That is, a clinical outcome is preferable to a biomarker. The use of biomarkers as surrogate end points is likely to be considered in settings in which clinical outcomes cannot be practically assessed; for example, in studies that examine prevention of Alzheimer’s disease in which the clinical outcomes (such as the time of onset of clinical Alzheimer’s disease) may not occur for many years after treatment initiation.

neurochemical markers from CSF that have high sensitivity and specificity for Alzheimer’s disease may be considered helpful for diagnostic purposes, whereas addition of a neuroimaging tool such as mRI or amyloidPeT may also offer the possibility to assess the effects of a particular treatment in these populations. multicentre and worldwide activities such as those carried out by the ADnI — which use several prespecified neurochemical and neuroimaging biomarkers, and standardization across sites and platforms98,99 — should help in further qualification of at least some of these biomarkers.

Creating innovation in biomarker development. Regulatory bodies such as the FDA and the emA have identified development of biomarkers as a high priority in general and particularly in dementia. To foster innovation in this field, consortia like the Biomarker Consortium, the Critical Path Initiative and the Innovative medicines Initiative, which involve collaboration of the different stakeholders, are highly welcomed and proactively supported. We think that such public–private partnerships (BOX 4) encourage detection and development of new biomarkers and improve consensus on the requirements of their validation process. This will be supported by the ongoing dialogue on harmonization of regulatory practices between the emA and the FDA, and both organizations will consider the experience with biomarkers in other fields — for example, oncology or cardiovascular medicine — for Alzheimer’s disease. Therefore regulators encourage the different stakeholders to involve regulatory bodies as early on as possible in biomarker development to accomplish better coordination of regulatory requirements and ongoing research in the different stages of drug development.

Box 2 | Qualification of a surrogate end point

Listed below are some aspects that need to be addressed when qualifying a biomarker as a surrogate end point:• Plausible connection between basic science and

clinical trials

• Is there a strong, independent, consistent association between surrogate end point and clinical outcome (necessary, not sufficient)?

• Evidence from randomized clinical trials that improvements in the surrogate end point leads consistently to improvement of the target outcome

• Large, precise, and lasting treatment effects

• Are the likely benefits worth the potential harms and costs?

Box 3 | regulatory view: potential use of biomarkers in clinical trials

Cerebrospinal fluid markers (for example, increased phosphophorylated tau and decreased amyloid-β

1-42 levels) are helpful as trait markers of Alzheimer’s disease

that have high sensitivity and specificity. However, they have yet to show value as markers of disease state.

Brain imaging (for example, magnetic resonance imaging (MRI) of the medial temporal lobe) can be helpful as trait markers for enrichment of populations at risk of developing Alzheimer’s disease, and serial MRI can be helpful as a marker of disease state. Brain imaging can also be used as an end point in dose-finding or proof-of-concept studies, and as a secondary end point in pivotal studies. Other brain imaging techniques such as positron emission tomography (PET) of amyloid-β or regional glucose (18F-2-fluoro-2-deoxy-d-glucose; FDG) metabolism can also be helpful as a trait marker. FDG-PET is potentially useful as a marker of disease state in proof-of-concept studies and as a secondary end point in pivotal studies.

R E V I E W S



Discussion and perspectivesThere is a consensus between academic institutions, industry and regulatory authorities that there are three potential uses of biomarkers in clinical trials. one use is as diagnostic tools to enrich the sample of patients with Alzheimer’s disease and to exclude other causes of dementia for which symptoms resemble Alzheimer’s disease. However, this could restrict the clinical indication of the investigational compound in the general population. Furthermore, as drug candidates with a diseasemodifying effect can be expected to be most effective in the earlier stages of the disease, biomarkers would be essential for early disease detection.

A second use of biomarkers is to identify and monitor the biochemical effect of a drug to facilitate drug development. This would increase the predictive power of nonimaging biomarkers when translating drug effects from mouse models of Alzheimer’s disease to the clinical setting. of the more than 40 molecules that reduce plaque burden in these animal models1 several lack preventive or clinical effect in treating patients with Alzheimer’s disease. In addition, small, shortterm clinical trials using

biomarkers can verify a biochemical effect of diseasemodifying drug candidates, before the expensive and timeconsuming step is taken to larger Phase II or III clinical trials. Indeed, biomarkers have already been implemented in Phase I and II studies in patients with Alzheimer’s disease, and are noncontroversial for regulators. But before a biomarker can be used as surrogate end points in Phase III studies, a link between the treatmentinduced change in the biomarker and the desired clinical outcome has to be established. largescale worldwide multicentre initiatives, such as the ADnI100,101, will support validation and qualification efforts in this respect.

A third application of biomarkers in clinical trials is to enable early and specific detection of side effects of the drug. As an example, in Aβ immunotherapy trials there is a risk of encephalitis10 and vasogenic oedema102, which may be difficult to identify clinically, but are easily detected by mRI and CSF analyses. Furthermore, if longitudinal mRI and CSF measures are part of a clinical trial, this can provide definitive information that a drug candidate is not associated with this type of adverse events.

Although several mRI, PeT and CSFbased biomarkers have proved useful as diagnostic markers in clinical trials, there are only some reports suggesting that biomarkers may identify and monitor the effect of the drug on the neurodegenerative process70,79,82, and there are no qualified surrogates for the clinical end point. It is hoped that the involvement of regulatory authorities as early on as possible in the biomarker discovery process will accomplish better coordination of regulatory requirements for the qualification of biomarkers, and that biomarker consortia, with strong collaboration of the different stakeholders, will foster the ongoing dialogue on harmonization of regulatory practices between the emA and the FDA.

Another challenge is the lack of drugs with an established diseasemodifying effect that can be used to validate that a certain biomarker will correlate with the biochemical effect of the drug and therefore predict the clinical outcome. At the same time, evidence that a drug candidate affects the central disease process and the characteristic neuropathology will be required to label the drug as diseasemodifying. This dual dependency introduces a catch22like situation in Alzheimer’s disease drug development and biomarker research. nevertheless, there are numerous ongoing clinical trials on diseasemodifying drug candidates that include biomarkers as end points. These trials will successively provide accumulating evidence on whether biomarkers will be useful as proofofconcept tools for the intended mechanism of action of a compound, as surrogate end points to predict the clinical outcome, and as the basis for a diseasemodifying claim of the drug.

Box 4 | Public–private partnerships

Like any scientific endeavour that involves multiple disciplines, biomarker development requires a high degree of collaboration between stakeholders so that basic research can be translated into clinical decisions. Although incremental progress is possible within individual organizations, this will be limited by resources and the intellectual capital possessed by these organizations. Furthermore, while an organization may be capable of investing significant resources towards the development of a particular biomarker in a given pathology or area of research, it is not likely to obtain a return on its investment if there is no complementary diagnostic product for Alzheimer’s disease on the market. Biomarker development that is independent of marketable products is not a feasible or sustainable business model.

Therefore, significant advances in this field are more likely to be achieved through joint investments and pooling of resources and expertise, with the payback being sharing outcomes and results, such as clinical trial data, that will stimulate competition in the marketplace and product development. There are many examples of strategic alliances that span multiple industries that were established with different goals103. The Biomarker Consortium is an example in which government agencies, academic institutions, industry, patient advocacy and professional organizations have assembled to bridge gaps in biomarker development. Public-health gaps in several disease areas are addressed including oncology, neuroscience, metabolic disorders, and immunity and inflammation. The Alzheimer’s disease Neuroimaging Initiative (ADNI) is a public–private partnership that tracks normal individuals, patients with mild cognitive impairment and patients with Alzheimer’s disease over a 5-year period. The ADNI aims to use the latest technologies and platforms to map out morphological changes in the human brain and to correlate them to disease progression. Analogous programmes have been initiated in Europe (the ADNI of the European Alzheimer’s disease Commission), Japan, and Germany (the Competence Network on Dementia). Ultimately, it is hoped that these studies will inform medical product development and bring safer, more effective and cheaper therapies to patients.

1. Blennow, K., de Leon, M. J. & Zetterberg, H. Alzheimer’s disease. Lancet 368, 387–403 (2006).

2. Klafki, H. W., Staufenbiel, M., Kornhuber, J. & Wiltfang, J. Therapeutic approaches to Alzheimer’s disease. Brain 129, 2840–2855, (2006).

3. Garcia-Alloza, M. et al. Existing plaques and neuritic abnormalities in APP:PS1 mice are not affected by administration of the gamma-secretase inhibitor LY-411575. Mol. Neurodegener. 4, 19 (2009).

4. Das, P., Murphy, M. P., Younkin, L. H., Younkin, S. G. & Golde, T. E. Reduced effectiveness of Aβ1–42 immuni zation in APP transgenic mice with significant amyloid deposition. Neurobiol. Aging 22, 721–727 (2001).

5. Levites, Y. et al. Anti-Aβ42- and anti-Aβ40-specific mAbs attenuate amyloid deposition in an Alzheimer disease mouse model. J. Clin. Invest. 116, 193–201 (2006).

6. Brookmeyer, R., Johnson, E., Ziegler-Graham, K., Arrighi, H. M. Forecasting the global burden of Alzhei-mer’s disease. Alzheimers Dement. 3, 186–191 (2007).

7. Siemers, E. R. How can we recognize “disease modification” effects? J. Nutr. Health Aging 13, 341–343 (2009).

8. Frank, R. & Hargreaves, R. Clinical biomarkers in drug discovery and development. Nature Rev. Drug Discov. 2, 566–580, (2003).

R E V I E W S



9. Frank, R. A. et al. Biological markers for therapeutic trials in Alzheimer’s disease. Proceedings of the biological markers working group; NIA initiative on neuroimaging in Alzheimer’s disease. Neurobiol. Aging 24, 521–536, (2003).

10. Orgogozo, J. M. et al. Subacute meningoencephalitis in a subset of patients with AD after Aβ42 immunization. Neurology 61, 46–54 (2003).

11. Blennow, K. & Hampel, H. CSF markers for incipient Alzheimer’s disease. Lancet Neurol. 2, 605–613 (2003).

12. Teipel, S. J., Meindl, T., Grinberg, L., Heinsen, H. & Hampel, H. Novel MRI techniques in the assessment of dementia. Eur. J. Nucl. Med. Mol. Imaging 35 (Suppl. 1), 58–69 (2008).

13. Hampel, H. et al. Core candidate neurochemical and imaging biomarkers of Alzheimer’s disease. Alzheimers Dement. 4, 38–48 (2008).

14. Whiting, P., Rutjes, A. W., Reitsma, J. B., Bossuyt, P. M. & Kleijnen, J. The development of QUADAS: a tool for the quality assessment of studies of diagnostic accuracy included in systematic reviews. BMC Med. Res. Methodol. 3, 25 (2003).

15. Nagy, Z. et al. Hippocampal pathology reflects memory deficit and brain imaging measurements in Alzheimer’s disease: clinicopathologic correlations using three sets of pathologic diagnostic criteria. Dementia 7, 76–81 (1996).

16. Logothetis, N. K. The neural basis of the blood-oxygen-level-dependent functional magnetic resonance imaging signal. Philos. Trans. R. Soc. Lond. B Biol. Sci. 357, 1003–1037 (2002).

17. Reivich, M. et al. The [18F]fluorodeoxyglucose method for the measurement of local cerebral glucose utilization in man. Circ. Res. 44, 127–137 (1979).

18. Lockhart, A. et al. PIB is a non-specific imaging marker of amyloid-β (Aβ) peptide-related cerebral amyloidosis. Brain 130, 2607–2615 (2007).

19. Allegri, R. F., Glaser, F. B., Taragano, F. E. & Buschke, H. Mild cognitive impairment: believe it or not? Int. Rev. Psychiatry 20, 357–363 (2008).

20. Fennema-Notestine, C. et al. Structural neuroimaging in the detection and prognosis of pre-clinical and early AD. Behav. Neurol. 21, 3–12 (2009).

21. Giedd, J. N. et al. Reliability of cerebral measures in repeated examinations with magnetic resonance imaging. Psychiatry Res. 61, 113–119 (1995).

22. Ewers, M. et al. Multicenter assessment of reliability of cranial MRI. Neurobiol. Aging 27, 1051–1059 (2006).

23. Bokde, A. L. et al. Decreased activation along the dorsal visual pathway after a 3-month treatment with galantamine in mild Alzheimer disease: a functional magnetic resonance imaging study. J. Clin. Psychopharmacol. 29, 147–156 (2009).

24. Moffett, J. R., Ross, B., Arun, P., Madhavarao, C. N. & Namboodiri, A. M. N-Acetylaspartate in the CNS: from neurodiagnostics to neurobiology. Prog. Neurobiol. 81, 89–131 (2007).

25. Kantarci, K. 1H magnetic resonance spectroscopy in dementia. Br. J. Radiol. 80, S146–S152 (2007).

26. Alexander, G. E., Chen, K., Pietrini, P., Rapoport, S. I. & Reiman, E. M. Longitudinal PET evaluation of cerebral metabolic decline in dementia: a potential outcome measure in Alzheimer’s disease treatment studies. Am. J. Psychiatry 159, 738–745 (2002).

27. Hirono, N., Hashimoto, M., Ishii, K., Kazui, H. & Mori, E. One-year change in cerebral glucose metabolism in patients with Alzheimer’s disease. J. Neuropsychiatry Clin. Neurosci. 16, 488–492 (2004).

28. Heiss, W. D., Kessler, J., Mielke, R., Szelies, B. & Herholz, K. Long-term effects of phosphatidylserine, pyritinol, and cognitive training in Alzheimer’s disease. A neuropsychological, EEG, and PET investigation. Dementia 5, 88–98 (1994).

29. Mega, M. S. et al. Metabolic patterns associated with the clinical response to galantamine therapy: a fludeoxyglucose F18 positron emission tomographic study. Arch. Neurol. 62, 721–728 (2005).

30. Stefanova, E. et al. Longitudinal PET evaluation of cerebral glucose metabolism in rivastigmine treated patients with mild Alzheimer’s disease. J. Neural Transm. 113, 205–218 (2006).

31. Kadir, A. et al. Effect of phenserine treatment on brain functional activity and amyloid in Alzheimer’s disease. Ann. Neurol. 63, 621–631 (2008).

32. Klunk, W. E. et al. Imaging brain amyloid in Alzheimer’s disease with Pittsburgh Compound-B. Ann. Neurol. 55, 306–319 (2004).

33. Jack, C. R. Jr et al. 11C PiB and structural MRI provide complementary information in imaging of Alzheimer’s disease and amnestic mild cognitive impairment. Brain 131, 665–680 (2008).

34. Maeda, J. et al. Longitudinal, quantitative assessment of amyloid, neuroinflammation, and anti-amyloid treatment in a living mouse model of Alzheimer’s disease enabled by positron emission tomography. J. Neurosci. 27, 10957–10968 (2007).

35. Villemagne, V. L. et al. Aβ deposits in older non-demented individuals with cognitive decline are indicative of preclinical Alzheimer’s disease. Neuropsychologia 46, 1688–1697 (2008).

36. Forsberg, A. et al. PET imaging of amyloid deposition in patients with mild cognitive impairment. Neurobiol. Aging 29, 1456–1465, (2008).

37. Rowe, C. C. et al. Imaging of amyloid beta in Alzheimer’s disease with 18F-BAY94-9172, a novel PET tracer: proof of mechanism. Lancet Neurol. 7, 129–135 (2008).

38. Bobinski, M. et al. The histological validation of post mortem magnetic resonance imaging-determined hippocampal volume in Alzheimer’s disease. Neuroscience 95, 721–725 (2000).

39. Vellas, B. et al. Long-term follow-up of patients immunized with AN1792: reduced functional decline in antibody responders. Curr. Alzheimer Res. 6, 144–151 (2009).

40. Saumier, D. et al. Lessons learned in the use of volumetric MRI in therapeutic trials in Alzheimer’s disease: the ALZHEMED (Tramiprosate) experience. J. Nutr. Health Aging 13, 370–372 (2009).

41. Harold, D. et al. Genome-wide association study identifies variants at CLU and PICALM associated with Alzheimer’s disease. Nature Genet. 41, 1088–1093 (2009).

42. Lambert, J. C. et al. Genome-wide association study identifies variants at CLU and CR1 associated with Alzheimer’s disease. Nature Genet. 41, 1094–1099 (2009).

43. Klunk, W. E. et al. Imaging the pathology of Alzheimer’s disease: amyloid-imaging with positron emission tomography. Neuroimaging Clin. N. Am. 13, 781–789 (2003).

44. Chen, K. et al. Correlations between apolipoprotein E ε4 gene dose and whole brain atrophy rates. Am. J. Psychiatry 164, 916–921 (2007).

45. Scahill, R. I., Schott, J. M., Stevens, J. M., Rossor, M. N. & Fox, N. C. Mapping the evolution of regional atrophy in Alzheimer’s disease: unbiased analysis of fluid-registered serial MRI. Proc. Natl Acad. Sci. USA 99, 4703–4707 (2002).

46. Scheuner, D. et al. Secreted amyloid β-protein similar to that in the senile plaques of Alzheimer’s disease is increased in vivo by the presenilin 1 and 2 and APP mutations linked to familial Alzheimer’s disease. Nature Med. 2, 864–870 (1996).

47. Myers, A. J. et al. A survey of genetic human cortical gene expression. Nature Genet. 39, 1494–1499 (2007).

48. Kauwe, J. S. et al. Alzheimer’s disease risk variants show association with cerebrospinal fluid amyloid β. Neurogenetics 10, 13–17 (2009).

49. Kauwe, J. S. et al. Variation in MAPT is associated with cerebrospinal fluid tau levels in the presence of amyloid-β deposition. Proc. Natl Acad. Sci. USA 105, 8050–8054 (2008).

50. Melzer, D. et al. A genome-wide association study identifies protein quantitative trait loci (pQTLs). PLoS Genet. 4, e1000072 (2008).

51. Makeeva, O., Stepanov, V., Puzyrev, V., Goldstein, D. B. & Grossman, I. Global pharmacogenetics: genetic substructure of Eurasian populations and its effect on variants of drug-metabolizing enzymes. Pharmacogenomics 9, 847–868 (2008).

52. Farrer, L. A. et al. Statement on use of apolipoprotein E testing for Alzheimer disease. American College of Medical Genetics/American Society of Human Genetics Working Group on ApoE and Alzheimer disease. JAMA 274, 1627–1629 (1995).

53. Green, R. C. et al. Disclosure of APOE genotype for risk of Alzheimer’s disease. N. Engl. J. Med. 361, 245–254 (2009).

54. Herukka, S. K. et al. CSF Aβ42, Tau and phosphorylated Tau, APOE ε4 allele and MCI type in progressive MCI. Neurobiol. Aging 28, 507–514 (2007).

55. Ray, S. et al. Classification and prediction of clinical Alzheimer’s diagnosis based on plasma signaling proteins. Nature Med. 13, 1359–1362 (2007).

56. Irizarry, M. C. Biomarkers of Alzheimer disease in plasma. NeuroRx 1, 226–234 (2004).

57. Vanderstichele, H. et al. Standardization of measurement of beta-amyloid(1–42) in cerebrospinal fluid and plasma. Amyloid 7, 245–258 (2000).

58. Peskind, E. R. et al. Safety and acceptability of the research lumbar puncture. Alzheimer Dis. Assoc. Disord. 19, 220–225 (2005).

59. Blennow, K. Cerebrospinal fluid protein biomarkers for Alzheimer’s disease. NeuroRx 1, 213–225 (2004).

60. Strozyk, D., Blennow, K., White, L. R. & Launer, L. J. CSF Aβ 42 levels correlate with amyloid-neuropathology in a population-based autopsy study. Neurology 60, 652–656 (2003).

61. Fagan, A. M. et al. Inverse relation between in vivo amyloid imaging load and cerebrospinal fluid Aβ42 in humans. Ann. Neurol. 59, 512–519 (2006).

62. Buerger, K. et al. CSF phosphorylated tau protein correlates with neocortical neurofibrillary pathology in Alzheimer’s disease. Brain 129, 3035–3041 (2006).

63. Shaw, L. M. et al. Cerebrospinal fluid biomarker signature in Alzheimer’s disease neuroimaging initiative subjects. Ann. Neurol. 65, 403–413, (2009).

64. Visser, P. J. et al. Prevalence and prognostic value of CSF markers of Alzheimer’s disease pathology in patients with subjective cognitive impairment or mild cognitive impairment in the DESCRIPA study: a prospective cohort study. Lancet Neurol. 8, 619–627, (2009).

65. Mattsson, N. et al. CSF biomarkers and incipient Alzheimer disease in patients with mild cognitive impairment. JAMA 302, 385–393, (2009).

66. Andreasson, U., Portelius, E., Andersson, M. E., Blennow, K. & Zetterberg, H. Aspects of β-amyloid as a biomarker for Alzheimer’s disease. Biomarkers Med. 1, 59–78 (2007).

67. de Jong, D., Kremer, B. P., Olde Rikkert, M. G. & Verbeek, M. M. Current state and future directions of neurochemical biomarkers for Alzheimer’s disease. Clin. Chem. Lab. Med. 45, 1421–1434 (2007).

68. Zhong, Z. et al. Levels of β-secretase (BACE1) in cerebrospinal fluid as a predictor of risk in mild cognitive impairment. Arch. Gen. Psychiatry 64, 718–726 (2007).

69. Zetterberg, H. et al. Elevated cerebrospinal fluid BACE1 activity in incipient Alzheimer disease. Arch. Neurol. 65, 1102–1107 (2008).

70. Davidsson, P. et al. Differential increase in cerebrospinal fluid-acetylcholinesterase after treatment with acetylcholinesterase inhibitors in patients with Alzheimer’s disease. Neurosci. Lett. 300, 157–160 (2001).

71. von der Kammer, H. et al. Muscarinic acetylcholine receptors activate expression of the EGR gene family of transcription factors. J. Biol. Chem. 273, 14538–14544 (1998).

72. Blennow, K. et al. Longitudinal stability of CSF biomarkers in Alzheimer’s disease. Neurosci. Lett. 419, 18–22 (2007).

73. Zetterberg, H. et al. Intra-individual stability of CSF biomarkers for Alzheimer’s disease over two years. J. Alzheimers Dis. 12, 255–260 (2007).

74. Lanz, T. A., Hosley, J. D., Adams, W. J. & Merchant, K. M. Studies of Aβ pharmacodynamics in the brain, cerebrospinal fluid, and plasma in young (plaque-free) Tg2576 mice using the γ-secretase inhibitor N2-[(2S)-2-(3,5-difluorophenyl)-2-hydroxyethanoyl]-N1-[(7S)-5-methyl-6-oxo-6,7-dihydro-5H-dibenzo[b, d]azepin-7-yl]-L-alaninamide (LY-411575). J. Pharmacol. Exp. Ther. 309, 49–55 (2004).

75. Anderson, J. J. et al. Reductions in β-amyloid concentrations in vivo by the γ-secretase inhibitors BMS-289948 and BMS-299897. Biochem. Pharmacol. 69, 689–698 (2005).

76. Fleisher, A. S. et al. Phase 2 safety trial targeting amyloid beta production with a gamma-secretase inhibitor in Alzheimer disease. Arch. Neurol. 65, 1031–1038 (2008).

77. Sankaranarayanan, S. et al. First demonstration of cerebrospinal fluid and plasma Aβ lowering with oral administration of a β-site amyloid precursor protein-cleaving enzyme 1 inhibitor in nonhuman primates. J. Pharmacol. Exp. Ther. 328, 131–140 (2009).

78. Adlard, P. A. et al. Rapid restoration of cognition in Alzheimer’s transgenic mice with 8-hydroxy quinoline analogs is associated with decreased interstitial Aβ. Neuron 59, 43–55 (2008).

79. Lannfelt, L. et al. Safety, efficacy, and biomarker findings of PBT2 in targeting Aβ as a modifying therapy for Alzheimer’s disease: a phase IIa, double-blind, randomised, placebo-controlled trial. Lancet Neurol. 7, 779–786 (2008).

R E V I E W S



80. Hesse, C. et al. Transient increase in total tau but not phospho-tau in human cerebrospinal fluid after acute stroke. Neurosci. Lett. 297, 187–190 (2001).

81. Zetterberg, H. et al. Neurochemical aftermath of amateur boxing. Arch. Neurol. 63, 1277–1280 (2006).

82. Gilman, S. et al. Clinical effects of Aβ immunization (AN1792) in patients with AD in an interrupted trial. Neurology 64, 1553–1562 (2005).

83. Degerman Gunnarsson, M., Kilander, L., Basun, H. & Lannfelt, L. Reduction of phosphorylated tau during memantine treatment of Alzheimer’s disease. Dement. Geriatr. Cogn. Disord. 24, 247–252 (2007).

84. Slamon, D. J. et al. Studies of the HER-2/neu proto-oncogene in human breast and ovarian cancer. Science 244, 707–712 (1989).

85. Dubois, B. et al. Research criteria for the diagnosis of Alzheimer’s disease: revising the NINCDS-ADRDA criteria. Lancet Neurol. 6, 734–746 (2007).

86. Carrillo, M. C. et al. Early risk assessment for Alzheimer’s disease. Alzheimers Dement. 5, 182–196 (2009).

87. Coats, M. & Morris, J. C. Antecedent biomarkers of Alzheimer’s disease: the adult children study. J. Geriatr. Psychiatry Neurol. 18, 242–244 (2005).

88. Mintun, M. A. et al. [11C]PIB in a nondemented population: potential antecedent marker of Alzheimer disease. Neurology 67, 446–452 (2006).

89. Mani, R. B. The evaluation of disease modifying therapies in Alzheimer’s disease: a regulatory viewpoint. Stat. Med. 23, 305–314 (2004).

90. Pepe, M. S. et al. Phases of biomarker development for early detection of cancer. J. Natl Cancer Inst. 93, 1054–1061 (2001).

91. Leber, P. Slowing the progression of Alzheimer disease: methodologic issues. Alzheimer Dis. Assoc. Disord. 11 (Suppl. 5), 10–21; discussion 37–19 (1997).

92. Woodcock, J. & Woosley, R. The FDA critical path initiative and its influence on new drug development. Annu. Rev. Med. 59, 1–12 (2008).

93. Katz, R. Biomarkers and surrogate markers: an FDA perspective. NeuroRx 1, 189–195 (2004).

94. Temple, R. Are surrogate markers adequate to assess cardiovascular disease drugs? JAMA 282, 790–795 (1999).

95. Baker, S. G. & Kramer, B. S. A perfect correlate does not a surrogate make. BMC Med. Res. Methodol. 3, 16 (2003).

96. Fleming, T. R. & DeMets, D. L. Surrogate end points in clinical trials: are we being misled? Ann. Intern. Med. 125, 605–613 (1996).

97. Fox, N. C. et al. Effects of Aβeta immunization (AN1792) on MRI measures of cerebral volume in Alzheimer disease. Neurology 64, 1563–1572 (2005).

98. Jack, C. R. Jr et al. The Alzheimer’s Disease Neuroimaging Initiative (ADNI): MRI methods. J. Magn. Reson. Imaging 27, 685–691 (2008).

99. Mueller, S. G. et al. Ways toward an early diagnosis in Alzheimer’s disease: The Alzheimer’s Disease Neuroimaging Initiative (ADNI). Alzheimers Dement. 1, 55–66 (2005).

100. Williams, S. A., Slavin, D. E., Wagner, J. A. & Webster, C. J. A cost-effectiveness approach to the qualification and acceptance of biomarkers. Nature Rev. Drug Discov. 5, 897–902 (2006).

101. Gutman, S. & Kessler, L. G. The US Food and Drug Administration perspective on cancer biomarker development. Nature Rev. Cancer 6, 565–571 (2006).

102. Grundman, M. & Black, R. O3-04-05: clinical trials of bapineuzumab, a β-amyloid-targeted immunotherapy in patients with mild to moderate Alzheimer’s disease. Alzheimers Dement. 4, T166 (2008).

103. Sanhai, W. R., Spiegel, J. & Ferrari, M. A critical path approach to advance nanoengineered medical products. Drug Discov. Today Technol. 4, 35–41 (2007).

104. Ridha, B. H. et al. Application of automated medial temporal lobe atrophy scale to Alzheimer disease. Arch. Neurol. 64, 849–854 (2007).

105. Byrum, C. E. et al. Accuracy and reproducibility of brain and tissue volumes using a magnetic resonance segmentation method. Psychiatry Res. 67, 215–234 (1996).

106. Jack, C. R. Jr et al. Medial temporal atrophy on MRI in normal aging and very mild Alzheimer’s disease. Neurology 49, 786–794 (1997).

107. Wang, P. N., Lirng, J. F., Lin, K. N., Chang, F. C. & Liu, H. C. Prediction of Alzheimer’s disease in mild cognitive impairment: a prospective study in Taiwan. Neurobiol. Aging 27, 1797–1806, (2006).

108. Fox, N. C., Cousens, S., Scahill, R., Harvey, R. J. & Rossor, M. N. Using serial registered brain magnetic resonance imaging to measure disease progression in Alzheimer disease: power calculations and estimates of sample size to detect treatment effects. Arch. Neurol. 57, 339–344 (2000).

109. Loubinoux, I. et al. Within-session and between-session reproducibility of cerebral sensorimotor activation: a test–retest effect evidenced with functional magnetic resonance imaging. J. Cereb. Blood Flow Metab. 21, 592–607 (2001).

110. Kircher, T. T., Erb, M., Grodd, W. & Leube, D. T. Cortical activation during cholinesterase-inhibitor treatment in Alzheimer disease: preliminary findings from a pharmaco-fMRI study. Am. J. Geriatr. Psychiatry 13, 1006–1013 (2005).

111. Bokde, A. L. W. et al. Decreased activation along the dorsal visual pathway after a 3-month treatment with galantamine in mild Alzheimer disease. J. Clin. Psychopharmacol. 29, 147–156 (2009).

112. Goekoop, R., Scheltens, P., Barkhof, F. & Rombouts, S. A. Cholinergic challenge in Alzheimer patients and mild cognitive impairment differentially affects hippocampal activation — a pharmacological fMRI study. Brain 129, 141–157 (2006).

113. Chao, L. L. et al. Reduced medial temporal lobe N-acetylaspartate in cognitively impaired but nondemented patients. Neurology 64, 282–289 (2005).

114. Jessen, F. et al. Treatment monitoring and response prediction with proton MR spectroscopy in AD. Neurology 67, 528–530 (2006).

115. Herholz, K. et al. Comparability of FDG PET studies in probable Alzheimer’s disease. J. Nucl. Med. 34, 1460–1466 (1993).

116. Edison, P. et al. Amyloid, hypometabolism, and cognition in Alzheimer disease: an [11C]PIB and [18F]FDG PET study. Neurology 68, 501–508 (2007).

117. Heiss, W. D. et al. Effect of piracetam on cerebral glucose metabolism in Alzheimer’s disease as measured by positron emission tomography. J. Cereb. Blood Flow Metab. 8, 613–617 (1988).

118. Teipel, S. J. et al. Effects of donepezil on cortical metabolic response to activation during 18FDG-PET in Alzheimer’s disease: a double-blind cross-over trial. Psychopharmacology (Berl.) 187, 86–94 (2006).

119. Engler, H. et al. Two-year follow-up of amyloid deposition in patients with Alzheimer’s disease. Brain 129, 2856–2866 (2006).

120. Bohnen, N. I. et al. Degree of inhibition of cortical acetylcholinesterase activity and cognitive effects by donepezil treatment in Alzheimer’s disease. J. Neurol. Neurosurg. Psychiatry 76, 315–319 (2005).

121. Andreasen, N. et al. Cerebrospinal fluid β-amyloid(1–42) in Alzheimer disease: differences between early- and late-onset Alzheimer disease and stability during the course of disease. Arch. Neurol. 56, 673–680 (1999).

122. Olsson, A. et al. Simultaneous measurement of β-amyloid(1–42), total tau, and phosphorylated tau (Thr181) in cerebrospinal fluid by the xMAP technology. Clin. Chem. 51, 336–345 (2005).

123. Fukuyama, R. et al. Age-dependent change in the levels of Aβ40 and Aβ42 in cerebrospinal fluid from control subjects, and a decrease in the ratio of Aβ42 to Aβ40 level in cerebrospinal fluid from Alzheimer’s disease patients. Eur. Neurol. 43, 155–160 (2000).

124. Kanai, M. et al. Longitudinal study of cerebrospinal fluid levels of tau, Aβ1–40, and Aβeta1–42(43) in Alzheimer’s disease: a study in Japan. Ann. Neurol. 44, 17–26 (1998).

125. Lewczuk, P. et al. Neurochemical diagnosis of Alzheimer’s dementia by CSF Aβ42, Aβ42/Aβ40 ratio and total tau. Neurobiol. Aging 25, 273–281 (2004).

126. Hansson, O. et al. Prediction of Alzheimer’s disease using the CSF Aβ42/Aβ40 ratio in patients with mild cognitive impairment. Dement. Geriatr. Cogn. Disord. 23, 316–320 (2007).

127. Shoji, M. et al. Combination assay of CSF tau, Aβ1–40 and Aβ1–42(43) as a biochemical marker of Alzheimer’s disease. J. Neurol. Sci. 158, 134–140 (1998).

128. de Leon, M. J. et al. Longitudinal cerebrospinal fluid tau load increases in mild cognitive impairment. Neurosci. Lett. 333, 183–186 (2002).

129. de Leon, M. J. et al. Longitudinal CSF and MRI biomarkers improve the diagnosis of mild cognitive impairment. Neurobiol. Aging 27, 394–401 (2006).

130. Olsson, A. et al. Measurement of α- and β-secretase cleaved amyloid precursor protein in cerebrospinal fluid from Alzheimer patients. Exp. Neurol. 183, 74–80 (2003).

131. Verheijen, J. H. et al. Detection of a soluble form of BACE-1 in human cerebrospinal fluid by a sensitive activity assay. Clin. Chem. 52, 1168–1174 (2006).

132. Holsinger, R. M., Lee, J. S., Boyd, A., Masters, C. L. & Collins, S. J. CSF BACE1 activity is increased in CJD and Alzheimer disease versus [corrected] other dementias. Neurology 67, 710–712 (2006).

133. Holsinger, R. M., McLean, C. A., Collins, S. J., Masters, C. L. & Evin, G. Increased beta-secretase activity in cerebrospinal fluid of Alzheimer’s disease subjects. Ann. Neurol. 55, 898–899 (2004).

134. Georganopoulou, D. G. et al. Nanoparticle-based detection in cerebral spinal fluid of a soluble pathogenic biomarker for Alzheimer’s disease. Proc. Natl Acad. Sci. USA 102, 2273–2276 (2005).

135. Vanderstichele, H. et al. Analytical performance and clinical utility of the INNOTEST PHOSPHO-TAU(181P) assay for discrimination between Alzheimer’s disease and dementia with Lewy bodies. Clin. Chem. Lab. Med. 44, 1472–1480 (2006).

136. Vanmechelen, E. et al. Quantification of tau phosphorylated at threonine 181 in human cerebrospinal fluid: a sandwich ELISA with a synthetic phosphopeptide for standardization. Neurosci. Lett. 285, 49–52, (2000).

137. Kohnken, R. et al. Detection of tau phosphorylated at threonine 231 in cerebrospinal fluid of Alzheimer’s disease patients. Neurosci. Lett. 287, 187–190 (2000).

138. Hampel, H. et al. Measurement of phosphorylated tau epitopes in the differential diagnosis of Alzheimer disease: a comparative cerebrospinal fluid study. Arch. Gen. Psychiatry 61, 95–102 (2004).

AcknowledgementsThe authors would like to thank K. Duggan for technical assist-ance. H.H. acknowledges support by the Science Foundation Ireland Investigator Neuroimaging Grant Programme (08/IN.1/B1846).

Competing interests statementThe authors declare competing financial interests: see web version for details.

DaTabasesEntrez Gene: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=geneAPOE | APP | CLU | CR1 | MAPT | PICALM | PSEN1 | PSEN2OMIM: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=OMIMAlzheimer’s diseaseUniProtKB: http://www.uniprot.orgBACE1

FUrTHer iNFormaTioNADNI: http://www.adni-info.orgADNI clinical trial: http://www.alzheimers.org/clinicaltrials/fullrec.asp?PrimaryKey=208Alzheimer’s Association Research Roundtable: http://www.alz.org/professionals_and_researchers_research_roundtable.aspGuidance for Industry Qualifying for Pediatric Exclusivity Under Section 505A of the Federal Food, Drug, and Cosmetic Act: http://www.fda.gov/downloads/Drugs/Development ApprovalProcess/DevelopmentResources/UCM049924.pdfImproved Predictivity of Efficacy Evaluation — Brain Disorders: http://www.imi-europe.org/Lists/IMIPublicationDocuments/3_%20Improved%20Predictivity%20of%20Efficacy%20Evaluation%20-%20Brain%20Disorders.pdfQualification of novel methodologies for drug development — Guidance to applicants: http://www.ema.europa.eu/pdfs/human/biomarkers/7289408en.pdfThe Biomarkers Consortium: http://www.biomarkersconsortium.orgThe Critical Path Initiative:http://www.fda.gov/ScienceResearch/SpecialTopics/CriticalPathInitiative/ucm204289.htm

sUPPLemeNTarY iNFormaTioNSee online article: S1 (table)

ALL Links Are AcTive in The onLine PDf

R E V I E W S



http://www.nature.com/nrd/journal/v9/n7/box/nrd3115_audecl.html

http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=gene









http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=OMIM

http://www.ncbi.nlm.nih.gov/omim/104300

http://www.uniprot.org

http://www.uniprot.org/uniprot/Q2HJ40

http://www.adni-info.org

http://www.alzheimers.org/clinicaltrials/fullrec.asp?PrimaryKey=208

http://www.alzheimers.org/clinicaltrials/fullrec.asp?PrimaryKey=208














http://www.biomarkersconsortium.org



http://www.nature.com/nrd/journal/vaop/ncurrent/suppinfo/nrd3115.html

Biomarkers for Alzheimer’s disease, NRDD 2010

Documents

Transcript of Biomarkers for Alzheimer’s disease, NRDD 2010