Reviews and CommentaryFree Access

State Of The Art

Neuroimaging and the Search for a Cure for Alzheimer Disease

Jeffrey R. Petrella

Jeffrey R. Petrella

Published Online:Dec 1 2013https://doi.org/10.1148/radiol.13122503

Abstract

As radiologists, our role in the workup of the dementia patient has long been limited by the sensitivity of our imaging tools and lack of effective treatment options. Over the past 30 years, we have made tremendous strides in understanding the genetic, molecular, and cellular basis of Alzheimer disease (AD). We now know that the pathologic features of AD are present 1 to 2 decades prior to development of symptoms, though currently approved symptomatic therapies are administered much later in the disease course. The search for true disease-modifying therapy continues and many clinical trials are underway. Current outcome measures, based on cognitive tests, are relatively insensitive to pathologic disease progression, requiring long, expensive trials with large numbers of participants. Biomarkers, including neuroimaging, have great potential to increase the power of trials by matching imaging methodology with therapeutic mechanism. One of the most important advances over the past decade has been the development of in vivo imaging probes targeted to amyloid beta protein, and one agent is already available for clinical use. Additional advances include automated volumetric imaging methods to quantitate cerebral volume loss. Use of such techniques in small, early phase trials are expected to significantly increase the number and quality of candidate drugs for testing in larger trials. In addition to a critical role in trials, structural, molecular, and functional imaging techniques can give us a window on the etiology of AD and other neurodegenerative diseases. This combination of developments has potential to bring diagnostic radiology to the forefront in AD research, therapeutic trials, and patient care.

©RSNA, 2013

Learning Objectives:

After reading the article and taking the test, the reader will be able to:

■ Describe the pathologic hallmarks of Alzheimer disease
■ Explain the amyloid hypothesis and its role in helping us understand Alzheimer disease, as well as problems and challenges that remain
■ Describe the appropriate use of amyloid imaging in Alzheimer disease diagnosis
■ Explain which biomarkers are likely best suited for tracking the progression of Alzheimer disease
■ Recognize the imaging findings of early complications of new anti-amyloid therapies

Accreditation and Designation Statement

The RSNA is accredited by the Accreditation Council for Continuing Medical Education (ACCME) to provide continuing medical education for physicians. The RSNA designates this journal-based activity for a maximum of 1.0 AMA PRA Category 1 Credit^TM. Physicans should claim only the credit commensurate with the extent of their participation in the activity.

Disclosure Statement

The ACCME requires that the RSNA, as an accredited provider of CME, obtain signed disclosure statements from the authors, editors, and reviewers for this activity. For this journal-based CME activity, author disclosures are listed at the end of this article.

For the past 3 decades, the practice of radiology has changed at an unprecedented pace, with significant beneficial effects on conditions such as cancer, vascular disease, and trauma. However, despite considerable advances in research, the area of dementia remains a stagnant backwater of our daily radiology practice. We are able to rule out rare causes of dementia using imaging, such as normal pressure hydrocephalus, subdural hematoma, or an intracranial mass lesion, but we cannot rule in the most common cause of dementia, Alzheimer disease (AD), which has, until recently, remained a pathologic diagnosis. Diagnostic tools, as used in our everyday practice, such as computed tomography (CT) and magnetic resonance (MR) imaging, are not sufficiently sensitive in the detection of AD, nor specific in ruling it out. To compound matters, even if there were an ideal diagnostic test, there are still no effective disease-modifying treatments for AD, causing many, including third-party payers, to question the utility of an early diagnosis.

Despite the considerable limitations of our current diagnostic and therapeutic options, there remains an urgent need to develop treatments designed to significantly slow or halt the progression of AD, as well as a means of early identification of patients who may be candidates for such interventions. This article will describe the substantial progress made toward this goal and how neuroimaging is playing an increasingly important role.

A Future Public Health Crisis?

As the world population has continued to grow over the past several decades, recently surpassing 7 billion in 2011, there has been a dramatic shift in population demographics that may have a significant impact on public health. A recent report on world population growth demonstrates that, likely for the first time in history, the world population under age 5 years is decreasing and that over age 65 years is increasing (1). If this trend continues, by 2050, the population older than 65 years will outnumber the population younger than 5 years by two- to threefold (Fig 1). Because aging is the biggest risk factor for AD, the implication of this population shift is that the incidence of AD, now 30 million worldwide, will triple or quadruple by 2050, with most new cases occurring in the category greater than age 85 years and in lower- and middle-income countries (Fig 2) (2). Let us consider the economic consequences. If one considers the current cost of dementia care worldwide in terms of gross national products, it would already rank as the world’s 18th largest economy, costing approximately $600 billion per year (2). The rising incidence of dementia, if unchecked, could therefore result in a global health and economic crisis of unparalleled proportions.

Figure 1: World population aging. The graph illustrates growth in the world population under age 5 years (<5) and over age 65 years *(65*+). If this trend continues, by 2050, the population older than 65 years will outnumber the population younger than 5 years by two- to threefold. (Reprinted, with permission, from reference 1.)

Figure 2: Dementia–a global public health crisis: The incidence of AD, now approximately 30 million cases worldwide, will triple to quadruple by 2050, with most new cases occurring in those older than 85 years and in those in lower and middle income countries, as illustrated in the graph. (Reprinted, with permission, from reference 2.)

Progress in AD: From Seminal Discovery to Promising Diagnostic Biomarkers

The good news is that over the past century, we have made tremendous progress in understanding AD (3). Not much more than 100 years ago, the terms “senility” or “hardening of the arteries” were used to denote a condition that was considered an inevitable part of growing old, rather than a specific disease process. William Osler, one of the most renowned and respected physicians of his day, was infamously quoted, “As it can be maintained that all the great advances have come from men under 40, so the history of the world shows that a very large proportion of the evils may be traced to the sexagenarians—nearly all the great mistakes politically and socially, all of the worst poems, most of the bad pictures, a majority of the bad novels, not a few of the bad sermons and speeches” (4). Though clearly an exaggeration, Osler’s quote indicates the general acceptance that cognitive decline and dementia was a natural and inevitable part of the aging process in humans. This was supported by the long-held concept that the foundation of disease is based on evidence of gross or microscopic pathologic findings, and at that time, there was no known specific neuropathology underlying dementia.

Alzheimer Makes a Seminal Discovery

Then, in 1907 a German psychiatrist and neuropathologist named Alois Alzheimer published his autopsy findings in an unusual case of a 55-year-old woman with a 4-year history of progressive dementia (5). He described his microscopic findings in the brain as a “tangled bundle of fibrils” and “military foci resulting from the deposit of a unique substance” and went on to describe the senile plaques and neurofibrillary tangles that still are considered the pathologic hallmarks of what we today call AD (Fig 3) (6,7).

Figure 3a: The characteristic pathologic features of AD, originally described by **(a)** Alois Alzheimer (1864–1915) in 1907, still considered essential for neuropathologic diagnosis: **(b)** senile plaques and **(c)** neurofibrillary tangles. **(b)** Note typical appearance of neocortical plaques with staining (hematoxylin-eosin stain; original magnification, ×200) and increased cellularity around the plaques, which consists primarily of reactive astrocytes. **(c)** Note typical appearance of a neurofibrillary tangle in a pyramidal neuron of the hippocampus (hematoxylin-eosin stain; original magnification, ×600). The tangle (arrow) appears as a circumscribed inclusion that extends from the cell body into the apical dendrite. (Reprinted, with permission, from reference 101.)

Figure 3b: The characteristic pathologic features of AD, originally described by **(a)** Alois Alzheimer (1864–1915) in 1907, still considered essential for neuropathologic diagnosis: **(b)** senile plaques and **(c)** neurofibrillary tangles. **(b)** Note typical appearance of neocortical plaques with staining (hematoxylin-eosin stain; original magnification, ×200) and increased cellularity around the plaques, which consists primarily of reactive astrocytes. **(c)** Note typical appearance of a neurofibrillary tangle in a pyramidal neuron of the hippocampus (hematoxylin-eosin stain; original magnification, ×600). The tangle (arrow) appears as a circumscribed inclusion that extends from the cell body into the apical dendrite. (Reprinted, with permission, from reference 101.)

Figure 3c: The characteristic pathologic features of AD, originally described by **(a)** Alois Alzheimer (1864–1915) in 1907, still considered essential for neuropathologic diagnosis: **(b)** senile plaques and **(c)** neurofibrillary tangles. **(b)** Note typical appearance of neocortical plaques with staining (hematoxylin-eosin stain; original magnification, ×200) and increased cellularity around the plaques, which consists primarily of reactive astrocytes. **(c)** Note typical appearance of a neurofibrillary tangle in a pyramidal neuron of the hippocampus (hematoxylin-eosin stain; original magnification, ×600). The tangle (arrow) appears as a circumscribed inclusion that extends from the cell body into the apical dendrite. (Reprinted, with permission, from reference 101.)

The Prevalence of AD Is Recognized

For almost 70 years there was hardly a mention of Alzheimer’s work in the medical literature, as it was considered a rare cause of dementia, with onset before age 65 years. Finally, in 1976, an American neurologist, Robert Katzmann, suggested a link between senility and AD, suggesting that many of the cases of senile dementia were pathologically identical to AD (8). If this were so, then AD was not a rare condition of middle age, but in fact, the fourth or fifth leading cause of death in the United States. Key officials at the National Institute on Aging decided at that time that AD was deserving of an organized national research effort.

The Pathologic Findings Are Characterized

In the 1980s the proteins underlying the plaques and tangles of AD were isolated, with amyloid beta (Aβ) protein underlying plaques, and tau protein underlying neurofibrillary tangles (9–11). By performing autopsy studies on patients at different clinical stages of AD, researchers began to understand how the disease spread throughout the brain, in a predictable pattern (12,13) (Fig 4). For example, in patients at the mildest stages of AD, neurofibrillary tangles were found only in a portion of the medial temporal lobe, the entorhinal cortex. This structure sends efferent fibers to the hippocampus, critical for encoding new memories. At progressively later stages, tangle pathology would be found in the hippocampal region, leading first to memory loss and then, as the disease progressed, to other areas of the temporal lobe and association cortices, leading to more widespread cognitive deficits, including language, attention, and executive function. In the late stages of the disease there is sparing only of the primary motor and visual cortices, which finally become involved in end-stage disease as patients lose activities of daily living and become bed-bound and completely dependent on external care. These were the first clues that we could understand the cognitive deficits in AD by understanding the functional anatomy of the brain regions affected by the underlying neuropathology. Such studies have led to different theories on what causes AD.

Figure 4a: Medial surface view of the cerebrum demonstrates pathologic progression of amyloid (left column) and neurofibrillary tangle (right column) deposits in various stages of AD, based on autopsy studies in the 1990s. **(a)** Darker shading represents greater amyloid deposition: In stage A, there is amyloid deposition in the basal frontal and temporal lobes. In stage B, there are amyloid deposits throughout all cortical association areas between the frontal, parietal, temporal, and occipital cortices, with only mild involvement of the hippocampus and sparing of the sensorimotor areas. In stage C, amyloid deposition extends throughout the entire cerebral cortex. **(b)** Neurofibrillary tangle deposition: Stages I and II show neurofibrillary deposition virtually confined to the medial temporal lobe, in a single layer of the entorhinal cortical region. Stages III and IV demonstrate more severe changes in the entorhinal cortical region as well involvement of additional limbic structures, such as the hippocampus. Stages V and VI are characterized by more extensive involvement of the cortex. (Reprinted, with permission, from reference 13.)

Figure 4b: Medial surface view of the cerebrum demonstrates pathologic progression of amyloid (left column) and neurofibrillary tangle (right column) deposits in various stages of AD, based on autopsy studies in the 1990s. **(a)** Darker shading represents greater amyloid deposition: In stage A, there is amyloid deposition in the basal frontal and temporal lobes. In stage B, there are amyloid deposits throughout all cortical association areas between the frontal, parietal, temporal, and occipital cortices, with only mild involvement of the hippocampus and sparing of the sensorimotor areas. In stage C, amyloid deposition extends throughout the entire cerebral cortex. **(b)** Neurofibrillary tangle deposition: Stages I and II show neurofibrillary deposition virtually confined to the medial temporal lobe, in a single layer of the entorhinal cortical region. Stages III and IV demonstrate more severe changes in the entorhinal cortical region as well involvement of additional limbic structures, such as the hippocampus. Stages V and VI are characterized by more extensive involvement of the cortex. (Reprinted, with permission, from reference 13.)

The Amyloid Hypothesis

Though it was clear that the presence of neuropathology at autopsy correlated with clinical manifestations of disease, evidence providing a causal link was still lacking. In vitro and in vivo experiments in the late 1980s and 1990s suggested that overproduction of Aβ protein may have a causal role in AD (14). The first genetic mutations related to AD were discovered in the gene for amyloid precursor protein (APP), a key component of Aβ metabolism (15–17). This gene was found to be located on chromosome 21, the same chromosome implicated in patients with Down syndrome, who also invariably express AD pathology (18). Other genes related to AD, the Presenilins, were subsequently found to play a role in enhancing the processing of APP metabolism leading to Aβ plaque formation (19–21). Experiments in knockout mice carrying Presenilin mutations demonstrated overproduction of Aβ plaques in the brains of these animals (22). Such evidence led to what is known as the amyloid hypothesis (23,24), which today remains the predominant theory for the cause of AD. The amyloid hypothesis posits that abnormal accumulation of Aβ in the brain is the primary influence in AD pathogenesis. The rest of the process, including tangle deposition and neurodegeneration, are considered downstream effects of an imbalance between Aβ production and clearance (14). Since this time, multiple theories linking amyloid and tau pathology have been proposed (25–27), including those based on recent biomarker studies, outlined below (28,29).

The First Therapies Are Developed

Also in the 1990s, the first therapies for AD were developed. These were based on the cholinergic hypothesis, which maintains that some of the most marked degenerative changes in AD occur in the cholinergic cells of the basal forebrain and decreased cholinergic transmission plays a major role in the cognitive and functional symptoms of AD (30,31). The first class of drugs, the cholinesterase inhibitors, was focused on improving the cognitive symptoms of AD by increasing neurotransmitter levels of acetylcholine in the brain (32). The first of these drugs to be approved by the U.S. Food and Drug Administration (FDA) was tacrine (33). Subsequently, a number of other better-tolerated cholinesterase inhibitors and a glutamate receptor antagonist were approved (34) and still remain the only FDA-approved treatments for AD to date. These drugs are considered symptomatic, rather than disease modifying, and therefore do not significantly affect the pathologic burden of disease; within a year or two, patients continue to decline cognitively, often approaching their original premedication trajectory. In the new millennium, the goal has been to develop a true disease-modifying therapy. There are currently more than 50 agents in ongoing, multicenter clinical trials, and they are summarized by mechanism in Table 1. The largest proportion of these therapies target the Aβ protein.

**Table 1** Classes of Disease-Modifying Therapies for AD Currently in Clinical Trials

Source.—Reference 104.

Amyloid Imaging Is a Major Advance of the Past Decade

Despite these advances in genetics, molecular biology, and therapeutics, there is considerable consensus in the AD research community that one of the most significant advances in AD research over the past decade has been in imaging—namely, the development of positron emission tomography (PET) radioligands that bind in vivo to fibrillar amyloid, the form of amyloid which is found in plaques. The first agent, termed Pittsburgh compound B (PIB), was developed at the University of Pittsburgh, based on a carbon 11 (¹¹C)-tagged analog of Congo red, the histologic stain for amyloid. Initial use of this compound in humans demonstrated a significant difference in cortical uptake and washout between AD patients and normal subjects (35), and subsequent studies using this agent have helped us make seminal advances in our understanding of how the presence of fibrillar amyloid relates to multiple genetic, clinical, and imaging variables in the living human brain (36–39).

Unfortunately, because of the 20-minute half-life of the ¹¹C-radioligand, this agent could only be used at facilities housing a cyclotron on its premises, thereby limiting potential widespread clinical availability of this agent. Since this time, a number of other amyloid-binding radiopharmaceuticals have been developed, namely those tagged to fluorine 18 (¹⁸F). Because of the 110-minute half-life of these agents, and the already established central-distribution infrastructure of ¹⁸F, it is estimated that 97% of the U.S. population will have access to these agents (40). Three such agents—florbetapir, flubetaben and flumetanol—are in late-stage clinical trials in humans. Florbetapir, recently approved by the FDA, is now clinically available for the indication of detecting Aβ fibrillar pathology in patients being evaluated for progressive cognitive decline (41,42). A recent study of this agent in end-of-life patients verified by means of autopsy that this tracer accurately targets amyloid plaques (43), suggesting that we are now able to see, in vivo, the pathology that Alois Alzheimer first described under the microscope at autopsy (Fig 5). Tau tracers for PET have also been tested in humans (44,45), and tracer uptake correlates with the specific type of memory impairment seen in AD (46), though the agents themselves also bind to amyloid plaques. The development of a tau-specific tracer that binds to neurofibrillary tangles, and not amyloid plaques, is underway, and preclinical work in animal models and human brain specimens has shown considerable promise (47). This ability to image amyloid plaques and neurofibrillary tangles in living humans has long been considered a “holy grail” of AD diagnosis and thus represents a major area of progress for AD research.

Figure 5: Validation of a radioligand that binds to fibrillar amyloid in vivo and shows great potential for early detection. Florbetapir, recently approved by the FDA, is now clinically available for the indication of detecting Aβ fibrillar pathology in patients being evaluated for progressive cognitive decline (41,42). Images show this tracer accurately targets amyloid plaques in humans. Left and middle columns show sagittal and axial florbetapir PET images in three subjects with variable levels of radiotracer uptake. Right column depicts Aβ antibody reactivity by immunohistochemistry in postmortem specimens from the same subjects. Note correlation between radiotracer uptake and antibody reactivity. (Reprinted, with permission, from reference 43.)

Problems with the Amyloid Hypothesis

Unfortunately, there are a number of unsettled issues with the amyloid hypothesis. First, 20%–40% of cognitively normal elderly have significant plaque deposition, despite being cognitively normal—it is not yet known if all such individuals will develop AD or not. Second, amyloid accumulation does not parallel cognitive decline, particularly in the late stages of the disease, and the correlation between amyloid and cognition even in earlier stages is modest, accounting for less than 20% of variability. Longitudinal data on patients with serial amyloid imaging and MR imaging studies suggests that amyloid deposition tends to plateau early, in the presymptomatic or prodromal phases of AD (48). Patients with AD tend not to accumulate significant additional amyloid over time, as measured with PET, yet continue to undergo cognitive decline and atrophy, reflecting continued neurodegeneration in the face of little additional amyloid accumulation (Fig 6). This suggests that amyloid imaging may be a good biomarker for early detection of preclinical AD or very mild AD, rather than as a means of tracking AD progression in more advanced cases.

Figure 6: A proposed model relating imaging pathology and cognition over an adult lifetime, based on analysis of data from two ongoing longitudinal imaging data registries (48). Horizontal axis = the lifetime clinical course. Neurodegeneration, detected with MR imaging, is indicated by a dashed line. Cognition is indicated by a dot-dash line. Amyloid deposition, detected by means of PIB, is indicated by a solid line late in life (portion of the disease for which there is data). Two possible theoretical trajectories (dotted lines), reflecting uncertainty about the time course of early amyloid deposition. Note that amyloid deposition tends to plateau early, in presymptomatic or prodromal phases of AD and does not parallel cognitive decline, particularly in late stages. Patients with AD tend not to accumulate significant additional amyloid over time, yet continue to undergo cognitive decline and neurodegenerative atrophy. (Reprinted, with permission, from reference 48.)

The exact role of fibrillar amyloid deposition in AD pathology is still unclear. Recent studies in familial AD cases (49) have shown that Aβ protein is detectable in the cerebrospinal fluid (CSF) as much as 25 years prior to symptomatic onset, and fibrillar amyloid changes on imaging are seen approximately 15 years before predicted symptom onset (Fig 7). However, tau changes in the CSF, indicative of neurodegeneration, are also occurring at approximately 15 years prior to onset. Such observations raise questions about the primacy of Aβ deposition in the AD pathophysiologic cascade. Last, a number of large multicenter phase III trials of antiamyloid therapies have failed, to date, and some have worsened patients owing to side effects (50). While active and passive immunotherapy directed against amyloid plaques has demonstrated a reduction in amyloid plaques in clinical trials (Fig 8) (51), there has been no evidence of corresponding effects in slowing cognitive decline or reducing atrophy, with the exception of a single agent, solanezumab, recently demonstrating a very modest effect on cognitive and functional decline in a subgroup of patients with mild AD (52). Further, there is no evidence, to date, that patients showing the greatest antiamyloid effects show the greatest cognitive benefits. These considerations have led scientists to revisit the amyloid hypothesis. Several prominent investigators have even suggested that amyloid plaques may merely be a byproduct of the disease process, or perhaps even a protective mechanism, rather than a causative agent itself (53,54).

Figure 7: Data from the Dominantly Inherited Alzheimer’s Network study illustrating amyloid changes at imaging years prior to symptom onset. A, Images compare fibrillar Aβ deposition, measured at PET with use of PIB, of the average of autosomal-dominant AD mutation carriers and noncarriers 20 years before estimated time of symptom onset. Note no significant differences in Aβ deposition. B, Images compare carriers and noncarriers 10 years before expected symptom onset. Note significantly more Aβ deposition in deep gray matter and cortex in mutation carriers vs noncarriers. C, Additional Aβ deposition seen in carriers throughout the cortex and deep gray structures at the estimated time of symptom onset. Increased standardized uptake value ratio *(SUVR)* indicates increased binding of PIB to fibrillar amyloid. The scale ranges from low SUVR values (blue), indicating low amounts of amyloid, to high SUVR values (red), indicating high amounts of amyloid. Line plot (right) of cross-sectional data demonstrates mean SUVR in the precuneus, one of the earliest sites of amyloid deposition, as a function of years from estimated symptom onset in carriers (red line) vs noncarriers (blue line). Compared with noncarriers, amyloid deposition in the precuneus begins approximately 15–20 years before expected symptom onset in carriers. Dashed lines = 95% confidence intervals of the fitted curves. (Reprinted, with permission, from reference 49.)

Figure 8: Images illustrate results of a phase II double-blind, placebo-controlled trial of bapineuzumab, a humanized anti-Aβ monoclonal antibody, which shows a treatment-related reduction in plaque burden in the brain. Trial used ¹¹C-PIB PET amyloid imaging to assess cortical amyloid load and measured change in PIB retention from baseline to week 78 as the primary outcome. Representative subjects from, A, bapineuzumab and, B, placebo groups show baseline and week 78 PET scans side by side. Color scale = PIB uptake ratios relative to cerebellum. Mean ¹¹C-PIB PET changes are shown at top center of each panel for each patient. C, Patients treated with bapineuzumab had a statistically significant reduction in mean tracer uptake compared with placebo group at 78 weeks = –0.24 (P = .003). (Reprinted, with permission, from reference 51.)

Proponents of the amyloid theory have maintained that antiamyloid therapies may not have worked because therapy was administered too late, in mild to moderate AD patients, after significant irreversible neurodegeneration and cognitive decline has already taken place (55). Because amyloid deposition appears to be an early event in the pathophysiology of AD, therapies directed at clearing plaques may be more effective earlier on. There are numerous ongoing and planned trials of antiamyloid therapies in patients with mild cognitive impairment (MCI) and asymptomatic individuals at risk for AD. Three, large, publically funded trials are detailed in Table 2. If trials targeting amyloid plaques continue to fail, it may be that the insoluble plaques themselves may be the wrong target. Indeed, it has been suggested that the soluble, oligomeric form of amyloid is the species directly toxic to neurons (14,56). It remains to be seen if these revisions of the amyloid theory will be supported by future data. Thus, though it appears that amyloid deposition is an early event in AD pathophysiology, its exact role is still unclear. Moreover, it is likely that a combination of biomarkers, including those that measure amyloid as well as downstream neurodegenerative effects, will be necessary for monitoring the entire disease course.

**Table 2** National Institute on Aging Funded Trials of Anti-amyloid Therapies

Source.—Reference 104.

Measures of Brain Atrophy

Because amyloid imaging may be less sensitive in tracking AD disease progression, particularly in the symptomatic phase, recent consensus statements have emphasized the importance of a biomarker of neurodegeneration, a critical component of AD pathophysiology in the prodromal and early dementia phases (29). Brain atrophy, particularly of the medial temporal lobe, where neurofibrillary tangle pathology is first seen, is considered an early marker of neurodegeneration. Volume-based MR imaging measures to quantitate cerebral volume loss in a variety of critical brain structures correlate well with cognitive decline as well as with tau pathology, suggesting that they might serve as accurate surrogate markers of neurodegenerative pathology (57).

Unfortunately, although developed and incorporated into clinical trials more than 20 years ago (58), these techniques have not played a significant role in patient care. This is primarily because of their labor-intensive nature, requiring manual or semimanual tracing of complex anatomic structures such as the hippocampus and amygdala. The recent development of completely automated, and therefore highly reproducible, volumetric methods is changing this (59,60). Such methods allow automated measurements of cortical thickness and subcortical volumes on high-resolution T1-weighted images across the entire brain that can be used to track disease progression over time. Regional changes can be assessed as a color overlay on brain images (Fig 9) or as quantitative region-of-interest measures in millimeters for cortical thickness or milliliters for volume. Versions of such automated methods are now commercially available for use in a clinical setting, allowing calculation of normalized percentile scores, adjusted for age, in individual patients, as well as volumetric changes in the same individual over time (Fig 10). While the technique is commercially available, full adoption awaits establishment of appropriate contexts of use and standards for diagnosis (61), as will be discussed later in this article.

Figure 9: Image depicts example of automated segmentation of the cortex and subcortical structures in an individual subject displayed as a multicolor overlay on a coronal T1-weighted image (top left panel) and left lateral surface rendering of the brain (bottom left panel). Changes in cortical thickness, and in hippocampal and ventricular volume, over 6- (center panels) and 12- (right panels) month periods are shown. Blue = volume or thickness decrease and red = volume or thickness increase over time. (Courtesy of Drs Dominic Holland and James Brewer, UCSD: Reprinted, with permission, from reference 103.)

Figure 10: Example of a commercial implementation (NeuroQuant) of automated volumetric brain measures. A, Multicolor overlay of segmented cortex and subcortical structures on orthogonal multiplanar T1-weighted images. Table shows raw volume, percentage of intracranial volume, and population normalized percentiles of volumetric measures in a normal control subject. B, Age-referenced plots of multiple volumetric measurements in an AD subject over time in the hippocampi (left) and lateral ventricles (right). As the patient ages, the rates of hippocampal volume decrease and ventricular volume increase exceed expected rates of changes due to normal aging. *ICV* = intracranial volume. (Image courtesy of Michael Kapp, Invivo, Pewaukee, Wis.)

The Multinational ADNI Biomarker Trial

These and other biomarkers, including fluorodeoxyglucose (FDG) PET and CSF proteins, are under intense investigation in the Alzheimer’s Disease Neuroimaging Initiative (ADNI), a multicenter, multinational natural history study across five continents, including the Americas, Europe, Australia, and Asia (62). Funded by a partnership among government, industry, and private philanthropic organizations, ADNI was initially intended to study the natural history of imaging and other biomarkers in AD, MCI, and control subjects. Such information was to form the basis for future clinical trials of therapeutics, using biomarkers as surrogate outcome measures. However, the significance of ADNI has gone considerably beyond its original scope. For example, because of the great effort devoted to protocol standardization and harmonization across multiple centers, devices, and time, this study represents a new standard for the performance of longitudinal multicenter imaging studies. Raw and processed data from ADNI is stored in a publically available database, and is accessible to any interested investigator throughout the world with a scientific question relevant to its data content (https://ida.loni.usc.edu/login.jsp).

Since its inception in 2003, there have been over 250 publications from the original 5-year ADNI study and its 2-year (ADNI-GO) and 5-year (ADNI-2) extensions. These are summarized in detail elsewhere (62). By elucidating the trajectories of imaging and other biomarkers in AD, MCI, and controls subjects, these data have significantly improved our understanding of the biology and pathobiology of aging, MCI, and AD. This has led to the development of early detection methods and the conclusion that the best diagnostic and prognostic classifiers combine optimum features from multiple modalities, including PET, MR imaging, CSF proteins, and cognitive measures (63–72) (Fig 11). This is not surprising, given the multiple, complex interacting factors constituting cognition and etiologies leading to cognitive decline.

Figure 11: Receiver operating characteristic curves for various models combining CSF, MR imaging, and FDG-PET biomarkers for predicting cognitive decline in subjects at risk for AD. Greater area under the curve *(AUC)* denotes higher diagnostic efficacy. Of the three biomarkers alone, FDG-PET added the most prognostic information with an AUC of 0.874, as compared with MR imaging (AUC = 0.741) and CSF proteins (AUC = 0.695). Note that highest area is produced by model combining all modalities (AUC = 0.921). Covariates represent demographic and cognitive variables and ApoE4 gene status. (Reprinted, with permission, from reference 69.)

Data from ADNI and other biomarker studies have led to a hypothetical model of the AD pathologic cascade (Fig 12) (28,29). In this model, the earliest sentinel event is considered an abnormality in Aβ processing, for example abnormal Aβ production, aggregation, and/or clearance, eventually leading to plaque deposition in the brain. After a lag time, amyloid plaque deposition is followed by neuronal metabolic and synaptic dysfunction, neurodegeneration, and finally, progressive cognitive symptoms. The model suggests that by assessing the level and trajectory of biomarkers of amyloid deposition, neuronal dysfunction and neurodegeneration, the progression of AD can be assessed even in its presymptomatic stage. See Table 3 for biomarkers under investigation in AD. Though the exact timing of biomarker abnormalities may be modified by genetic and environmental factors, such as Apo E genotype and cardiovascular risk factors, the temporal sequence of biomarker abnormalities should remain fixed. This model, developed for the more common, late-onset or sporadic AD, has been supported by data in early onset autosomal dominant AD mutation carriers (49).

Figure 12: Hypothetical model of AD pathologic cascade, characterized by biomarker data. Aβ protein is deposited initially, while the patient is asymptomatic, and may be measured by amyloid markers in the CSF or by amyloid PET imaging (red curve). Neuronal injury subsequently occurs and may be measured by tau markers in the CSF or by FDG-PET (blue curve). Neuronal injury is followed by neuronal death, and may be measured by atrophic changes on MR imaging (light green curve). Patients then become symptomatic, their memory first affected (purple curve). This MCI phase is followed by deficits in other domains, leading to impaired daily function (dark green curve), dementia, and, eventually, total dependence. (Reprinted, with permission, from reference 29.)

**Table 3** Biomarkers under Investigation in AD

Note.—Adapted, with permission, from reference 73. BOLD = blood oxygen level-dependent, SPECT = single photon emission tomography.

Redefinition of AD

Such biomarkers have led to a redefinition of AD for the first time in 27 years, with an emphasis on earlier diagnosis at the prodromal (MCI) and even preclinical phases, before significant irreversible neurodegeneration has taken place (Fig 13). In the traditional conceptualization, the term “Alzheimer disease” has been reserved only for patients with a clinical dementia syndrome, that is, “Alzheimer disease” was equivalent to “Alzheimer dementia.” Under the current definition, there is recognition that the pathology of AD is present even when patients are asymptomatic or minimally symptomatic, years and perhaps decades prior to the development of full-blown symptoms of dementia. This new definition emphasizes the concept that these earlier phases, termed MCI due to AD and preclinical AD, can be identified with the help of biomarkers, and therapy can be instituted earlier, possibly before significant irreversible damage has taken place. The designations “dementia due to AD” and “MCI due to AD” are considered new terms to be used in clinical practice (73,74). The designation “preclinical AD”, in which patients are asymptomatic, having only biomarker evidence of AD, is currently reserved for clinical trials only (75). Detailed criteria for dementia due to AD, MCI due to AD, and preclinical AD are shown in Table 4.

Figure 13a: Redefining AD. Schematic of the **(a)** traditional and **(b)** new conceptualization of AD. Declining cognitive function is depicted as a function of time. Under the traditional definition, the term “Alzheimer disease” was reserved for only for patients with a dementia syndrome. Under the new definition, there is recognition that the pathology of AD is present even when patients are asymptomatic or minimally symptomatic. This new definition emphasizes the concept that these earlier phases, termed MCI due to AD and preclinical AD, can be identified with the help of biomarkers, and therapy can be instituted earlier, possibly before significant irreversible damage has taken place. (Reprinted, with permission, from reference 75.)

Figure 13b: Redefining AD. Schematic of the **(a)** traditional and **(b)** new conceptualization of AD. Declining cognitive function is depicted as a function of time. Under the traditional definition, the term “Alzheimer disease” was reserved for only for patients with a dementia syndrome. Under the new definition, there is recognition that the pathology of AD is present even when patients are asymptomatic or minimally symptomatic. This new definition emphasizes the concept that these earlier phases, termed MCI due to AD and preclinical AD, can be identified with the help of biomarkers, and therapy can be instituted earlier, possibly before significant irreversible damage has taken place. (Reprinted, with permission, from reference 75.)

**Table 4** New Diagnostic Criteria, Incorporating Biomarkers, for AD and MCI and New Research Criteria for Preclinical AD

Note.—Adapted, with permission, from references 73–75.

Biomarkers and Clinical Trials of AD Therapeutics

With use of biomarkers and these new definitions, clinical trials enriched with patients at the preclinical and prodromal phases of AD may hasten the discovery of effective treatments to slow or halt the disease. Further, the placebo group in such trials may decline at a more homogeneous and faster rate if enriched for AD pathology. Multicenter phase III clinical trials, required by regulatory agencies prior to approval of a treatment for AD, are required to use outcome measures based on cognition, such as tests of memory or day-to-day function. Because, realistically, most disease-modifying therapies under development are expected to only show small effects on rates of change of these measures, such trials are heavily powered and expensive to carry out, perhaps requiring up to 1000 patients over 2 to 3 years to test a single agent for efficacy. Given the number of agents that have already failed at the phase III level (50), a better approach is needed for finding an effective treatment (55,76). A cliché among researchers in AD therapeutics is that the “cure” is probably in someone’s test tube today and it is merely a matter of finding it. The problem is that, although we have made significant advances in understanding the pathologic and pathophysiologic factors associated with AD, such as amyloid, tau, inflammation, oxidative stress, and neuronal vulnerability, we still lack a detailed understanding of the causal roles of these factors and interactions among them. As a result, there are many potential pathways to target, yet we lack precise knowledge as to which ones to target that would slow disease progression clinically. Thus, finding a cure will likely require high throughput techniques for dozens and perhaps hundreds of agents in earlier phase trials to select the appropriate agents to bring to the larger, more expensive, phase III trials.

Using a mechanistic imaging approach (Fig 14) in small clinical trials, matching a surrogate imaging marker with a particular drug mechanism, allows us to identify those agents that hit their target, even though the effects may be subclinical. For example, amyloid imaging can be paired to antiamyloid agents, volumetric imaging to trophic agents, such as nerve growth factor, and functional MR imaging to agents that modify cognitive brain networks, such as those involved in memory or attention (Fig 15) (77). Such an approach can markedly increase the power to detect a subclinical therapeutic effect and be used to select the most promising agents early on, before significant financial resources are invested in a large multicenter trial of a single agent. A considerable number of clinical trials of AD therapeutics using biomarkers have already been completed and are summarized elsewhere (78). It is important to remember that although an agent may hit its intended target, it may not necessarily have its intended therapeutic effect, as illustrated in a prior example (Fig 7); thus larger trials documenting clinical efficacy are ultimately required.

Figure 14: Demonstration of the concept of mechanistic imaging for use in clinical trials of new therapeutics in AD. Horizontal line represents a schematic of the pathophysiologic progression of AD from birth to through death. Arrows denote points in the pathophysiology where various imaging biomarkers may prove most sensitive. Pairing an imaging biomarker to the appropriate pathophysiologic events presents an opportunity for early diagnosis, prognosis, and as a means to monitor the effectiveness of a therapeutic to affect its intended target. *DTI* = diffusion tensor imaging, *fMRI* = functional MR imaging.

Figure 15a: **(a, b)** Example of a 12-week randomized, placebo-controlled trial of the cholinesterase inhibitor, donepezil, in subjects with MCI. The study incorporated functional MR imaging *(fMRI)* into the trial at a single center. **(a)** Images demonstrate thresholded group statistical activation map of entire group at baseline and drug and placebo arms posttreatment. Note increased activation in the right inferior frontal lobe (blue crosshairs) in the donepezil group compared with the placebo group. **(b)** Bar graph demonstrates mean activation magnitude in the right inferior frontal lobe at baseline and after therapy in both groups. There was a significant drug effect in the donepezil, but not the placebo, group. Of note, there were no significant drug effects detected by means of cognitive testing in the single center trial, though significant cognitive effects were demonstrated in the larger multicenter trial of which it was a part (102), supporting the role of imaging for detecting subclinical effects in smaller trials. (Reprinted, with permission, from reference 77.)

Figure 15b: **(a, b)** Example of a 12-week randomized, placebo-controlled trial of the cholinesterase inhibitor, donepezil, in subjects with MCI. The study incorporated functional MR imaging *(fMRI)* into the trial at a single center. **(a)** Images demonstrate thresholded group statistical activation map of entire group at baseline and drug and placebo arms posttreatment. Note increased activation in the right inferior frontal lobe (blue crosshairs) in the donepezil group compared with the placebo group. **(b)** Bar graph demonstrates mean activation magnitude in the right inferior frontal lobe at baseline and after therapy in both groups. There was a significant drug effect in the donepezil, but not the placebo, group. Of note, there were no significant drug effects detected by means of cognitive testing in the single center trial, though significant cognitive effects were demonstrated in the larger multicenter trial of which it was a part (102), supporting the role of imaging for detecting subclinical effects in smaller trials. (Reprinted, with permission, from reference 77.)

Radiologists’ Role in Clinical Care and Trials

Radiologists are uniquely suited to play a significant role in biomarker development, in treatment monitoring in clinical trials, and ultimately, in development of accurate diagnostic tests for evaluating patients who may be candidates for therapy. Perhaps the biggest question relates to whether there will soon be an early diagnostic test for AD. Given the lack of effective disease-modifying treatments, this question has remained in the background, but will become increasingly important as better therapies become available. The recent FDA approval of the amyloid-binding ¹⁸F radioligand, florbetapir (Amyvid) has ignited discussions regarding its potential for misuse outside its approved indication, primarily to rule out the likelihood that the patient’s cognitive impairment is due to AD (42). Professional societies, such as the Society for Nuclear Medicine, have drafted guidelines for appropriate use (79). One area of particular concern is whether amyloid PET scanning would be used as a screening tool to predict future disease in cognitively normal subjects with a family history of AD, or those with only subjective memory impairment (80), uses which are not yet supported by current evidence. Though evidence in MCI subjects supports that those in whom findings are positive for the amyloid tracer are more likely to decline compared with those in whom findings are negative (81–83) (Fig 16), the increasing incidence of amyloid positivity with age can confound interpretation in older patients. Thus, in individual subjects, an amyloid scan can still not be used to predict future decline with precision, though at a group level the effects are significant. At present, the primary application of florbetapir is to reduce the probability of AD or dementia with Lewy bodies if one has a negative scan, as well as applications in clinical trial settings for sample enrichment and to confirm target engagement.

Figure 16: Bar and whisker plot demonstrates change from baseline on cognitive assessment measures in subject with MCI over 18-month follow-up. Aβ positive subjects show significant (P < .05) declines on all cognitive measures, compared with Aβ negative subjects. On this graph, direction of change on the Alzheimer’s Disease Assessment Scale–Cognitive Subscale *(ADAS)* and Clinical Dementia Rating–Sum of Boxes *(CDR SB)* are reversed to show worsening scores as a negative change consistent with other measures. *DSS* = Digit Symbol Substitution, *MMSE* = Mini-Mental State Examination, *WMS-I* = Wechsler Memory Scale Immediate, *WMS-D* = Wechsler Memory Scale Delayed. (Reprinted, with permission, from reference 81.)

Studies of diagnostic tests for AD will need to provide appropriate evidence to guide third-party payer decisions, including estimates of test-retest reliability, diagnostic accuracy in community-based populations, impact on clinical decision making and treatment, as well as quality-of-life outcomes for patients and caregivers. This highlights not only the importance of sensitivity of biomarkers in clinical diagnosis, but also the importance of specificity. While abnormal hippocampal volume or amyloid imaging may suggest that a patient’s cognitive impairment is due to AD, the specificity of such changes, particularly in elderly patients in a community setting with multiple comorbidities, has not been established. To these ends, validated methods for binary reading of florbetapir PET images (as positive or negative) have been developed (43), and online training is available (http://AmyvidTraining.com). Similarly, efforts are underway to standardize and validate hippocampal volumetry as a biomarker for clinical diagnosis (61). Organizations such as the Institute for Clinical Economic Review at Harvard are working on models that would help define the specific type of evidence needed to guide third-party payer decisions.

Radiologists will also be needed for treatment monitoring in clinical trials. For example, several complications in clinical trials of antiamyloid immunotherapies have been identified with MR imaging, termed amyloid-related imaging abnormalities (ARIA). Because antiamyloid immunotherapies do not distinguish between amyloid deposits in blood vessels verses those in brain, treatment can lead to disruption of the blood-brain barrier, raising the risk for rare complications such as vasogenic edema or intracranial hemorrhage. Patients who are at high risk may potentially be identified with amyloid imaging prior to entry, given that cerebral amyloid angiopathy–related hemorrhages occur preferentially at sites of increased amyloid deposition (84). During therapy, subtle MR imaging signs may develop in patients who are asymptomatic within the first few infusions of the antiamyloid agents, including sulcal effusions (designated as ARIA-E) and microhemorrhages (designated as ARIA-H) (Fig 17) (85). By close monitoring of these findings it is hoped that rare complications of hemorrhage can be prevented by delaying dosing until the findings resolve. Radiologists can therefore play a pivotal role in safety monitoring during clinical trials by identifying these imaging findings in patients prior to selection or during the initial infusions of the agents.

Figure 17a: **(a, b)** Examples of ARIA seen on MR images used to monitor patients undergoing anti-amyloid immunotherapies. ARIA may manifest as edema, termed ARIA-E, here depicted as a focal area of sulcal hyperintensity on fluid-attenuated inversion recovery images **(a)**, and consistent with a sulcal effusion. It may also manifest as hemorrhage, termed ARIA-H, here depicted as sulcal hypointensity (arrows) on gradient-echo **(b)** or susceptibility-weighted images, and consistent with a sulcal hemorrhage. ARIA-E and ARIA-H can also affect the parenchyma. (Reprinted, with permission, from reference 85.)

Figure 17b: **(a, b)** Examples of ARIA seen on MR images used to monitor patients undergoing anti-amyloid immunotherapies. ARIA may manifest as edema, termed ARIA-E, here depicted as a focal area of sulcal hyperintensity on fluid-attenuated inversion recovery images **(a)**, and consistent with a sulcal effusion. It may also manifest as hemorrhage, termed ARIA-H, here depicted as sulcal hypointensity (arrows) on gradient-echo **(b)** or susceptibility-weighted images, and consistent with a sulcal hemorrhage. ARIA-E and ARIA-H can also affect the parenchyma. (Reprinted, with permission, from reference 85.)

Lastly, though most of the imaging done today is not quantitative, biomarker use in clinical trials will increasingly require quantitation. Imaging devices in use today have been optimized to create images suitable for visual interpretation and are not the accurately calibrated measuring instruments that will be required for clinical trials of the future. Along with several other organizations, the Radiological Society of North America is committed to the goal of quantitative imaging through support of the Quantitative Imaging Biomarkers Alliance, or QIBA, a group of stakeholders from academia, industry, and government regulatory agencies focused on improving the value and practicality of quantitative biomarkers by reducing variability across devices, patients, and time (86).

Beyond Clinical Care and Trials: Uncovering the Pathophysiology of AD and Other Dementias

Though the potential benefits of neuroimaging in AD are clear for drug development and early diagnosis, one of the most exciting aspects about its application to AD is that it may help unlock the mysteries underlying the disease itself, ultimately leading to prevention and cure. In the past this has been the exclusive purview of fields such as molecular and cellular biology and, more recently, genetics. However, the combination of structural, molecular, and functional imaging techniques may give us a window on the etiology not only of AD but also of other neurodegenerative diseases.

Amyloid and Brain Connections

As previously discussed, there is increasing evidence in animal models that the smaller, more soluble form of amyloid, rather the larger plaques demonstrated by amyloid-binding radiotracers, may play a pivotal role in early AD pathogenesis by disrupting synaptic function, leading to dysfunction in cell-to-cell communication between neurons (56,87). However, this theory is difficult to test in humans because assays for oligomers are not yet validated. It is possible that imaging technologies such as functional MR imaging may reveal disruption of brain connectivity as result of early amyloid pathology. In particular, failure of a unique brain network, termed the default mode network (88), has been implicated as playing a role in the cognitive deficits underlying AD (89–91). This network has gained increased attention over the past 5 years by both cognitive and clinical neuroscientists and is considered unique because it becomes less active during performance of a wide variety of cognitive tasks, including reading, actively listening, and recalling specific information, and becomes more active during rest. Evidence for the existence of this brain network is supported by resting-state FDG PET studies showing high metabolism in this collection of regions as well as resting-state functional MR imaging and electroencephalography studies demonstrating correlation of low-frequency spontaneous signal oscillations among these regions (92).

Several groups have studied AD, MCI, and controls subjects by using techniques such as functional MR imaging and functional connectivity MR imaging, demonstrating alteration in the activity of the major regions of the default mode network, as well as disruption in intra- and interregional connectivity in AD and MCI as well as cognitively normal subjects with amyloid deposition (89,91,93,94). Moreover, early breakdown in the connections in this brain network in MCI subjects correlates with future cognitive decline over a period of 2–3 years (90) (Fig 18). Recent studies have examined brain connectivity on a global and local scale by using graph theory techniques to investigate the topological network architecture of the entire brain in amnestic MCI (95). These studies have corroborated disruption of components of the default mode network as the major brain network alteration in amnestic MCI. In addition, global network function is less efficient, with longer paths of communication between functional brain areas and impaired communication between different functional brain networks. Importantly, these abnormal network metrics correlated with patients’ cognitive performance and distinguished them from controls with a high degree of sensitivity and specificity. This evidence suggests that measures of brain network function may play a role as early diagnostic, prognostic, and predictive biomarkers in AD clinical care and therapeutic trials.

Figure 18: Functional connectivity maps of the default mode network in normal elderly subjects (top row), subjects with MCI who later remained stable over 2-year follow-up (middle row), and subjects with MCI that later converted to AD over a 2-year follow-up. Subjects with MCI that later converted to AD over the follow-up period demonstrate significantly less activity in this network, compared with those MCI subjects who remained stable, and looked more like subjects with AD (not shown), whereas the stable MCI subjects looked more like the elderly controls (top row). Findings suggest that this network may represent a target for development of therapeutics that enhance its activity, and functional connectivity measures may play a role in early diagnosis. (Reprinted, with permission, from reference 90.)

Large-Scale Brain Networks and Neurodegenerative Disease

Studies of large-scale brain networks using functional, structural, and molecular imaging techniques may give us insight as to why certain proteins, such as amyloid, may target particular brain regions, and lead to a better understanding of the mechanisms underlying AD and other neurodegenerative disorders. Visual inspection of groupwise amyloid imaging maps in patients with AD and functional connectivity maps in young healthy subjects suggests that the pattern of amyloid deposition in the brain in patients with AD closely parallels that of the major regions of the default mode network in young healthy subjects, involving the medial parietal, frontal, and lateral parietal regions of association cortex (Fig 19). This observation has been extended to a hypothesis, termed the “overuse hypothesis,” linking overactivity in the default mode network in early life with amyloid deposition in later life (96). This theory postulates that during periods when the brain is awake, but not engaged in particular activity, the default mode network is most active. Overactivity, or overuse, of this network, for example due to lack of mental engagement, may lead to preferential amyloid deposition in these areas. This theory would explain why mental engagement, stimulating work, and strong social networks have been shown to protect against AD.

Figure 19: Images demonstrate the link between, A, pattern of functional brain connections in young healthy subjects, determined at resting-state functional MR imaging, and, C, Aβ deposition in patients with AD, determined at PET imaging with ¹¹C-PIB. Images demonstrate voxelwise statistical group maps of degree (A), a measure of functional connectivity, and Aβ deposition (C) in common stereotactic space. Note the areas of overlap in the medial parietal, frontal, and lateral parietal regions of association cortex, a set of regions commonly referred to as the default mode network. B, Scatterplot over stereotactic space demonstrates a high correlation between connectivity in young healthy volunteers and amyloid deposition in AD patients, suggesting that amyloid plaques may preferentially target highly connected brain regions such as the association cortices of the default mode network. (Reprinted, with permission, from reference 96.)

Disruption of large-scale brain networks may characterize AD as well as other neurodegenerative disorders. Striking similarities have been suggested between the topology of large-scale human brain networks in healthy individuals and syndrome-specific brain atrophy patterns in various neurodegenerative diseases including AD and various frontotemporal lobar dementia syndromes (97). These observations suggest that protein deposition disorders, such as AD and frontotemporal lobar dementia, selectively target large-scale human brain networks (Fig 20). Such work, which is based on structural and functional MR imaging studies, has been extended to test several theories explaining why neurodegenerative diseases might selectively target large-scale human brain networks (98). Recent data analyzed using network-based graph theoretical analysis suggests a transneuronal spread model of neurodegenerative pathology along networked axons by means of a mechanism similar to that proposed in prion diseases (99). This model has also been supported by recent data investigating the spread of tau protein from the entorhinal cortex in rodent models (100).

Figure 20: Image demonstrates the close relationship between syndrome-specific atrophy patterns in, A, patients and corresponding, B, functional and, C, structural brain networks in healthy controls. Syndrome-specific atrophy patterns determined by group differences between patients and healthy controls using voxel-based morphometry (A). Seed regions, denoted by circle, were used to generate intrinsic functional brain networks using resting-state functional functional MR imaging (B), and structural covariance (C) brain networks using covariance mapping of voxel-based morphometry measures across subjects. Note similar patterns within syndromes, suggesting that neurodegenerative disease, as seen in patients, may target particular brain networks in healthy subjects. Results are displayed on representative sections of the Montreal Neurologic Institute template brain. For visualization purposes, results are shown at P < .00001 uncorrected (A and C) and P < .001 corrected height and extent thresholds (B). Color bars indicate t scores, a measure of statistical significance based on the samples. On coronal and axial images, the left side of the image corresponds to the left side of the brain. *ANG* = angular gyrus, FI = frontoinsula, *IFG* = inferior frontal gyrus, *PMC* = premotor cortex, *TPole* = temporal pole, *bvFTD* = behavioral variant frontotemporal dementia, SD = semantic dementia, *PNFA* = progressive nonfluent aphasia, *CBS* = corticobasal syndrome. (Reprinted, with permission, from reference 97.)

Summary and Conclusion

As the world population ages, the incidence of dementia, particularly AD, is predicted to markedly increase, threatening to give rise to a global health and economic crisis of unparalleled proportions. There is urgency to find effective treatments that will halt or slow the progression of AD, delay its onset, or even prevent it altogether. The past 2 decades of research on AD have led to significant advances in our understanding of the genetic, molecular, and cellular basis of AD and to promising diagnostic biomarkers and therapies. Unfortunately, current therapies are symptomatic only, directed at slowing progression of cognitive decline for a limited time. The search for a true disease-modifying therapy continues and many clinical trials are underway.

Biomarkers, including neuroimaging, have great potential to increase the power of clinical trials through greater effect sizes by matching imaging methodology with therapeutic mechanism. The multicenter ADNI trial aims at identification and characterization of such biomarkers for clinical trials. In addition, it is highly likely that the combination of structural, molecular and functional imaging techniques can give us a window on etiology of not only AD but also other neurodegenerative diseases. This combination of developments has potential to bring diagnostic radiology to the forefront in AD research, therapeutic trials, and patient care.

Essentials

■ Alzheimer disease (AD) is poised to become a major public health problem, with societal and economic impact of global proportion.
■ Significant advances in our understanding of AD have led to new potential therapies and diagnostic markers, such as amyloid and volumetric imaging.
■ The multicenter Alzheimer’s Disease Neuroimaging Initiative and other biomarker trials are sorting out which markers are useful for clinical trials of therapeutic agents and have helped redefine AD for the first time in 27 years.
■ Radiologists can play a role in biomarker development, treatment monitoring in clinical trials, and ultimately, in development of accurate diagnostic tests for patient care.
■ Beyond clinical trials and patient care, imaging may help us unlock the mysteries of underlying AD and other neurodegenerative disorders.

Acknowledgments:

I would like to thank colleague and friend, Dr P. Murali Doraiswamy, for his thoughtful insights and critical feedback in developing this article, and Dr Burton P. Drayer, a leader and pioneer in the imaging of neurodegenerative disorders, for inviting me to deliver the RSNA 2011 Annual Oration in Diagnostic Radiology, on which the content of this article was based.

References

1. Haub C. World population aging: clocks illustrate growth in population under age 5 and over age 65. Population Bulletin. http://www.prb.org/Articles/2011/agingpopulationclocks.aspx. Published 2011.
Google Scholar
2. Wimo A, Prince M. World Alzheimer Report 2010: The Global Economic Impact of Dementia. London, England: Alzheimer’s Disease International, 2010.
Google Scholar
3. Petrella JR, Doraiswamy PM. Alzheimer’s disease: 100 years of progress. Neuroimaging Clin N Am 2005;15(4):xiii–xiv.
MedlineGoogle Scholar
4. Ballenger JF. Self, Senility, and Alzheimer’s Disease in Modern America: A History. Baltimore, Md: The Johns Hopkins University Press, 2006.
Google Scholar
5. Alzheimer A. Uber eine eigenartige Erkrankung der Hirnrinde. Allgemeine Zeitschrift fur Psychiatrie und phychish-Gerichtliche Medizin 1907;64:146–148.
Google Scholar
6. Alzheimer A. A unique illness involving the cerebral cortex. In: Rothenberg DA, Hochberg FH, eds. Neurological Classics in Modern Translation. London, England: Hafner, 1977;41–43.
Google Scholar
7. Hyman BT, Phelps CH, Beach TG, et al. National Institute on Aging-Alzheimer’s Association guidelines for the neuropathologic assessment of Alzheimer’s disease. Alzheimers Dement 2012;8(1):1–13.
MedlineGoogle Scholar
8. Katzman R. The prevalence and malignancy of Alzheimer disease: a major killer (editorial). Arch Neurol 1976;33(4):217–218.
MedlineGoogle Scholar
9. Grundke-Iqbal I, Iqbal K, Quinlan M, Tung YC, Zaidi MS, Wisniewski HM. Microtubule-associated protein tau: a component of Alzheimer paired helical filaments. J Biol Chem 1986;261(13):6084–6089.
MedlineGoogle Scholar
10. Kosik KS, Joachim CL, Selkoe DJ. Microtubule-associated protein tau (tau) is a major antigenic component of paired helical filaments in Alzheimer disease. Proc Natl Acad Sci U S A 1986;83(11):4044–4048.
MedlineGoogle Scholar
11. Masters CL, Simms G, Weinman NA, Multhaup G, McDonald BL, Beyreuther K. Amyloid plaque core protein in Alzheimer disease and Down syndrome. Proc Natl Acad Sci U S A 1985;82(12):4245–4249.
MedlineGoogle Scholar
12. Braak H, Braak E. Neuropathological stageing of Alzheimer-related changes. Acta Neuropathol (Berl) 1991;82(4):239–259.
MedlineGoogle Scholar
13. Braak H, Braak E. Frequency of stages of Alzheimer-related lesions in different age categories. Neurobiol Aging 1997;18(4):351–357.
MedlineGoogle Scholar
14. Hardy J, Selkoe DJ. The amyloid hypothesis of Alzheimer’s disease: progress and problems on the road to therapeutics. Science 2002;297(5580):353–356.
MedlineGoogle Scholar
15. Goldgaber D, Lerman MI, McBride OW, Saffiotti U, Gajdusek DC. Characterization and chromosomal localization of a cDNA encoding brain amyloid of Alzheimer’s disease. Science 1987;235(4791):877–880.
MedlineGoogle Scholar
16. Tanzi RE, Gusella JF, Watkins PC, et al. Amyloid beta protein gene: cDNA, mRNA distribution, and genetic linkage near the Alzheimer locus. Science 1987;235(4791):880–884.
MedlineGoogle Scholar
17. Robakis NK, Ramakrishna N, Wolfe G, Wisniewski HM. Molecular cloning and characterization of a cDNA encoding the cerebrovascular and the neuritic plaque amyloid peptides. Proc Natl Acad Sci U S A 1987;84(12):4190–4194.
MedlineGoogle Scholar
18. St George-Hyslop PH, Tanzi RE, Polinsky RJ, et al. The genetic defect causing familial Alzheimer’s disease maps on chromosome 21. Science 1987;235(4791):885–890.
MedlineGoogle Scholar
19. Borchelt DR, Thinakaran G, Eckman CB, et al. Familial Alzheimer’s disease-linked presenilin 1 variants elevate Abeta1-42/1-40 ratio in vitro and in vivo. Neuron 1996;17(5):1005–1013.
MedlineGoogle Scholar
20. Citron M, Westaway D, Xia W, et al. Mutant presenilins of Alzheimer’s disease increase production of 42-residue amyloid beta-protein in both transfected cells and transgenic mice. Nat Med 1997;3(1):67–72.
MedlineGoogle Scholar
21. Scheuner D, Eckman C, Jensen M, et al. Secreted amyloid beta-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. Nat Med 1996;2(8):864–870.
MedlineGoogle Scholar
22. Duff K, Eckman C, Zehr C, et al. Increased amyloid-beta42(43) in brains of mice expressing mutant presenilin 1. Nature 1996;383(6602):710–713.
MedlineGoogle Scholar
23. Hardy JA, Higgins GA. Alzheimer’s disease: the amyloid cascade hypothesis. Science 1992;256(5054):184–185.
MedlineGoogle Scholar
24. Selkoe DJ. The molecular pathology of Alzheimer’s disease. Neuron 1991;6(4):487–498.
MedlineGoogle Scholar
25. Dong S, Duan Y, Hu Y, Zhao Z. Advances in the pathogenesis of Alzheimer’s disease: a re-evaluation of amyloid cascade hypothesis. Transl Neurodegener 2012;1(1):18.
MedlineGoogle Scholar
26. Duan Y, Dong S, Gu F, Hu Y, Zhao Z. Advances in the pathogenesis of Alzheimer’s disease: focusing on tau-mediated neurodegeneration. Transl Neurodegener 2012;1(1):24.
MedlineGoogle Scholar
27. Hardy J. The relationship between amyloid and tau. J Mol Neurosci 2003;20(2):203–206.
MedlineGoogle Scholar
28. Jack CR Jr, Knopman DS, Jagust WJ, et al. Tracking pathophysiological processes in Alzheimer’s disease: an updated hypothetical model of dynamic biomarkers. Lancet Neurol 2013;12(2):207–216.
MedlineGoogle Scholar
29. Jack CR Jr, Knopman DS, Jagust WJ, et al. Hypothetical model of dynamic biomarkers of the Alzheimer’s pathological cascade. Lancet Neurol 2010;9(1):119–128.
MedlineGoogle Scholar
30. Bartus RT, Dean RL 3rd, Beer B, Lippa AS. The cholinergic hypothesis of geriatric memory dysfunction. Science 1982;217(4558):408–414.
MedlineGoogle Scholar
31. Francis PT, Palmer AM, Snape M, Wilcock GK. The cholinergic hypothesis of Alzheimer’s disease: a review of progress. J Neurol Neurosurg Psychiatry 1999;66(2):137–147.
MedlineGoogle Scholar
32. Krall WJ, Sramek JJ, Cutler NR. Cholinesterase inhibitors: a therapeutic strategy for Alzheimer disease. Ann Pharmacother 1999;33(4):441–450.
MedlineGoogle Scholar
33. Giacobini E. Invited review: cholinesterase inhibitors for Alzheimer’s disease therapy— from tacrine to future applications. Neurochem Int 1998;32(5-6):413–419.
MedlineGoogle Scholar
34. Lleó A, Greenberg SM, Growdon JH. Current pharmacotherapy for Alzheimer’s disease. Annu Rev Med 2006;57:513–533.
MedlineGoogle Scholar
35. Klunk WE, Engler H, Nordberg A, et al. Imaging brain amyloid in Alzheimer’s disease with Pittsburgh compound-B. Ann Neurol 2004;55(3):306–319.
MedlineGoogle Scholar
36. Fagan AM, Mintun MA, Mach RH, et al. Inverse relation between in vivo amyloid imaging load and cerebrospinal fluid Abeta42 in humans. Ann Neurol 2006;59(3):512–519.
MedlineGoogle Scholar
37. Mintun MA, Larossa GN, Sheline YI, et al. [11C]PIB in a nondemented population: potential antecedent marker of Alzheimer disease. Neurology 2006;67(3):446–452.
MedlineGoogle Scholar
38. Pike KE, Savage G, Villemagne VL, et al. Beta-amyloid imaging and memory in non-demented individuals: evidence for preclinical Alzheimer’s disease. Brain 2007;130(Pt 11):2837–2844.
MedlineGoogle Scholar
39. Rowe CC, Ng S, Ackermann U, et al. Imaging beta-amyloid burden in aging and dementia. Neurology 2007;68(20):1718–1725.
MedlineGoogle Scholar
40. Patterson JC 2nd, Mosley ML. How available is positron emission tomography in the United States? Mol Imaging Biol 2005;7(3):197–200.
MedlineGoogle Scholar
41. Seibyl J, Zubal IG, Jennings D, Marek K, Doraiswamy PM. Molecular PET imaging in multicenter Alzheimer’s therapeutic trials: current trends and implementation strategies. Expert Rev Neurother 2011;11(12):1783–1793.
MedlineGoogle Scholar
42. Yang L, Rieves D, Ganley C. Brain amyloid imaging: FDA approval of florbetapir F18 injection. N Engl J Med 2012;367(10):885–887.
MedlineGoogle Scholar
43. Clark CM, Schneider JA, Bedell BJ, et al. Use of florbetapir-PET for imaging beta-amyloid pathology. JAMA 2011;305(3):275–283.
MedlineGoogle Scholar
44. Shoghi-Jadid K, Small GW, Agdeppa ED, et al. Localization of neurofibrillary tangles and beta-amyloid plaques in the brains of living patients with Alzheimer disease. Am J Geriatr Psychiatry 2002;10(1):24–35.
MedlineGoogle Scholar
45. Small GW, Kepe V, Ercoli LM, et al. PET of brain amyloid and tau in mild cognitive impairment. N Engl J Med 2006;355(25):2652–2663.
MedlineGoogle Scholar
46. Tolboom N, van der Flier WM, Yaqub M, et al. Differential association of [11C]PIB and [18F]FDDNP binding with cognitive impairment. Neurology 2009;73(24):2079–2085.
MedlineGoogle Scholar
47. Fodero-Tavoletti MT, Okamura N, Furumoto S, et al. 18F-THK523: a novel in vivo tau imaging ligand for Alzheimer’s disease. Brain 2011;134(Pt 4):1089–1100.
MedlineGoogle Scholar
48. Jack CR Jr, Lowe VJ, Weigand SD, et al. Serial PIB and MRI in normal, mild cognitive impairment and Alzheimer’s disease: implications for sequence of pathological events in Alzheimer’s disease. Brain 2009;132(Pt 5):1355–1365.
MedlineGoogle Scholar
49. Bateman RJ, Xiong C, Benzinger TL, et al. Clinical and biomarker changes in dominantly inherited Alzheimer’s disease. N Engl J Med 2012;367(9):795–804.
MedlineGoogle Scholar
50. Haas C. Strategies, development, and pitfalls of therapeutic options for Alzheimer’s disease. J Alzheimers Dis 2012;28(2):241–281.
MedlineGoogle Scholar
51. Rinne JO, Brooks DJ, Rossor MN, et al. 11C-PiB PET assessment of change in fibrillar amyloid-beta load in patients with Alzheimer’s disease treated with bapineuzumab: a phase 2, double-blind, placebo-controlled, ascending-dose study. Lancet Neurol 2010;9(4):363–372.
MedlineGoogle Scholar
52. Doody RS. Phase 3 Studies of Solanezumab for Mild to Moderate Alzheimer’s Disease. In: Proceedings of the137th annual meeting of the American Neurological Association. Boston, Mass: American Neurological Association, 2012.
Google Scholar
53. Castellani RJ, Lee HG, Siedlak SL, et al. Reexamining Alzheimer’s disease: evidence for a protective role for amyloid-beta protein precursor and amyloid-beta. J Alzheimers Dis 2009;18(2):447–452.
MedlineGoogle Scholar
54. Lee HG, Casadesus G, Zhu X, Takeda A, Perry G, Smith MA. Challenging the amyloid cascade hypothesis: senile plaques and amyloid-beta as protective adaptations to Alzheimer disease. Ann N Y Acad Sci 2004;1019:1–4.
MedlineGoogle Scholar
55. Sperling RA, Jack CR Jr, Aisen PS. Testing the right target and right drug at the right stage. Sci Transl Med 2011;3(111):111cm33.
Google Scholar
56. Selkoe DJ. Soluble oligomers of the amyloid beta-protein impair synaptic plasticity and behavior. Behav Brain Res 2008;192(1):106–113.
MedlineGoogle Scholar
57. Whitwell JL, Josephs KA, Murray ME, et al. MRI correlates of neurofibrillary tangle pathology at autopsy: a voxel-based morphometry study. Neurology 2008;71(10):743–749.
MedlineGoogle Scholar
58. Jack CR Jr, Bentley MD, Twomey CK, Zinsmeister AR. MR imaging-based volume measurements of the hippocampal formation and anterior temporal lobe: validation studies. Radiology 1990;176(1):205–209.
Google Scholar
59. Fischl B, Salat DH, Busa E, et al. Whole brain segmentation: automated labeling of neuroanatomical structures in the human brain. Neuron 2002;33(3):341–355.
MedlineGoogle Scholar
60. Fischl B, van der Kouwe A, Destrieux C, et al. Automatically parcellating the human cerebral cortex. Cereb Cortex 2004;14(1):11–22.
MedlineGoogle Scholar
61. Jack CR Jr, Barkhof F, Bernstein MA, et al. Steps to standardization and validation of hippocampal volumetry as a biomarker in clinical trials and diagnostic criterion for Alzheimer’s disease. Alzheimers Dement 2011;7(4):474, e4.
MedlineGoogle Scholar
62. Carrillo MC, Bain LJ, Frisoni GB, Weiner MW. Worldwide Alzheimer’s disease neuroimaging initiative. Alzheimers Dement 2012;8(4):337–342.
MedlineGoogle Scholar
63. Cui Y, Liu B, Luo S, et al. Identification of conversion from mild cognitive impairment to Alzheimer’s disease using multivariate predictors. PLoS ONE 2011;6(7):e21896.
MedlineGoogle Scholar
64. Davatzikos C, Bhatt P, Shaw LM, Batmanghelich KN, Trojanowski JQ. Prediction of MCI to AD conversion, via MRI, CSF biomarkers, and pattern classification. Neurobiol Aging 2011;32(12):e19–e27.
MedlineGoogle Scholar
65. Ewers M, Walsh C, Trojanowski JQ, et al. Prediction of conversion from mild cognitive impairment to Alzheimer’s disease dementia based upon biomarkers and neuropsychological test performance. Neurobiol Aging 2012;33(7):1203–1214.
MedlineGoogle Scholar
66. Gomar JJ, Bobes-Bascaran MT, Conejero-Goldberg C, Davies P, Goldberg TE; Alzheimer’s Disease Neuroimaging Initiative. Utility of combinations of biomarkers, cognitive markers, and risk factors to predict conversion from mild cognitive impairment to Alzheimer disease in patients in the Alzheimer’s disease neuroimaging initiative. Arch Gen Psychiatry 2011;68(9):961–969.
MedlineGoogle Scholar
67. Hinrichs C, Singh V, Mukherjee L, et al. Spatially augmented LPboosting for AD classification with evaluations on the ADNI dataset. Neuroimage 2009;48(1):138–149.
MedlineGoogle Scholar
68. Kohannim O, Hua X, Hibar DP, et al. Boosting power for clinical trials using classifiers based on multiple biomarkers. Neurobiol Aging 2010;31(8):1429–1442.
MedlineGoogle Scholar
69. Shaffer JL, Petrella JR, Sheldon FC, et al. Predicting cognitive decline in subjects at risk for Alzheimer disease by using combined cerebrospinal fluid, MR imaging, and PET biomarkers. Radiology 2013;266(2):583–591.
Google Scholar
70. Vemuri P, Wiste HJ, Weigand SD, et al. MRI and CSF biomarkers in normal, MCI, and AD subjects: predicting future clinical change. Neurology 2009;73(4):294–301.
MedlineGoogle Scholar
71. Vemuri P, Wiste HJ, Weigand SD, et al. MRI and CSF biomarkers in normal, MCI, and AD subjects: diagnostic discrimination and cognitive correlations. Neurology 2009;73(4):287–293.
MedlineGoogle Scholar
72. Walhovd KB, Fjell AM, Brewer J, et al. Combining MR imaging, positron-emission tomography, and CSF biomarkers in the diagnosis and prognosis of Alzheimer disease. AJNR Am J Neuroradiol 2010;31(2):347–354.
MedlineGoogle Scholar
73. Albert MS, DeKosky ST, Dickson D, et al. The diagnosis of mild cognitive impairment due to Alzheimer’s disease: recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimers Dement 2011;7(3):270–279.
MedlineGoogle Scholar
74. McKhann GM, Knopman DS, Chertkow H, et al. The diagnosis of dementia due to Alzheimer’s disease: recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimers Dement 2011;7(3):263–269.
MedlineGoogle Scholar
75. Sperling RA, Aisen PS, Beckett LA, et al. Toward defining the preclinical stages of Alzheimer’s disease: recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimers Dement 2011;7(3):280–292.
MedlineGoogle Scholar
76. Selkoe DJ. Resolving controversies on the path to Alzheimer’s therapeutics. Nat Med 2011;17(9):1060–1065.
MedlineGoogle Scholar
77. Petrella JR, Prince SE, Krishnan S, Husn H, Kelley L, Doraiswamy PM. Effects of donepezil on cortical activation in mild cognitive impairment: a pilot double-blind placebo-controlled trial using functional MR imaging. AJNR Am J Neuroradiol 2009;30(2):411–416.
MedlineGoogle Scholar
78. Dickerson BC, Sperling RA. Neuroimaging biomarkers for clinical trials of disease-modifying therapies in Alzheimer’s disease. NeuroRx 2005;2(2):348–360.
MedlineGoogle Scholar
79. Johnson KA, Minoshima S, Bohnen NI, et al. Appropriate use criteria for amyloid PET: a report of the Amyloid Imaging Task Force, the Society of Nuclear Medicine and Molecular Imaging, and the Alzheimer’s Association. J Nucl Med 2013;54(3):476–490.
MedlineGoogle Scholar
80. Moghbel MC, Saboury B, Basu S, et al. Amyloid-β imaging with PET in Alzheimer’s disease: is it feasible with current radiotracers and technologies? Eur J Nucl Med Mol Imaging 2012;39(2):202–208.
MedlineGoogle Scholar
81. Doraiswamy PM, Sperling RA, Coleman RE, et al. Amyloid-β assessed by florbetapir F 18 PET and 18-month cognitive decline: a multicenter study. Neurology 2012;79(16):1636–1644.
MedlineGoogle Scholar
82. Forsberg A, Engler H, Almkvist O, et al. PET imaging of amyloid deposition in patients with mild cognitive impairment. Neurobiol Aging 2008;29(10):1456–1465.
MedlineGoogle Scholar
83. Okello A, Koivunen J, Edison P, et al. Conversion of amyloid positive and negative MCI to AD over 3 years: an 11C-PIB PET study. Neurology 2009;73(10):754–760.
MedlineGoogle Scholar
84. Gurol ME, Dierksen G, Betensky R, et al. Predicting sites of new hemorrhage with amyloid imaging in cerebral amyloid angiopathy. Neurology 2012;79(4):320–326.
MedlineGoogle Scholar
85. Sperling RA, Jack CR Jr, Black SE, et al. Amyloid-related imaging abnormalities in amyloid-modifying therapeutic trials: recommendations from the Alzheimer’s Association Research Roundtable Workgroup. Alzheimers Dement 2011;7(4):367–385.
MedlineGoogle Scholar
86. Buckler AJ, Bresolin L, Dunnick NR, Sullivan DC; Group. A collaborative enterprise for multi-stakeholder participation in the advancement of quantitative imaging. Radiology 2011;258(3):906–914.
Google Scholar
87. Selkoe DJ. Alzheimer’s disease is a synaptic failure. Science 2002;298(5594):789–791.
MedlineGoogle Scholar
88. Raichle ME, MacLeod AM, Snyder AZ, Powers WJ, Gusnard DA, Shulman GL. A default mode of brain function. Proc Natl Acad Sci U S A 2001;98(2):676–682.
MedlineGoogle Scholar
89. Hedden T, Van Dijk KR, Becker JA, et al. Disruption of functional connectivity in clinically normal older adults harboring amyloid burden. J Neurosci 2009;29(40):12686–12694.
MedlineGoogle Scholar
90. Petrella JR, Sheldon FC, Prince SE, Calhoun VD, Doraiswamy PM. Default mode network connectivity in stable vs progressive mild cognitive impairment. Neurology 2011;76(6):511–517.
MedlineGoogle Scholar
91. Sperling RA, Laviolette PS, O’Keefe K, et al. Amyloid deposition is associated with impaired default network function in older persons without dementia. Neuron 2009;63(2):178–188.
MedlineGoogle Scholar
92. Buckner RL, Andrews-Hanna JR, Schacter DL. The brain’s default network: anatomy, function, and relevance to disease. Ann N Y Acad Sci 2008;1124:1–38.
MedlineGoogle Scholar
93. Petrella JR, Wang L, Krishnan S, et al. Cortical deactivation in mild cognitive impairment: high-field-strength functional MR imaging. Radiology 2007;245(1):224–235. [Published correction appears in Radiology 2008;246(1):338.]
Google Scholar
94. Sheline YI, Raichle ME, Snyder AZ, et al. Amyloid plaques disrupt resting state default mode network connectivity in cognitively normal elderly. Biol Psychiatry 2010;67(6):584–587.
MedlineGoogle Scholar
95. Wang J, Zuo X, Dai Z, et al. Disrupted functional brain connectome in individuals at risk for Alzheimer’s disease. Biol Psychiatry 2013;73(5):472–481.
MedlineGoogle Scholar
96. Buckner RL, Snyder AZ, Shannon BJ, et al. Molecular, structural, and functional characterization of Alzheimer’s disease: evidence for a relationship between default activity, amyloid, and memory. J Neurosci 2005;25(34):7709–7717.
MedlineGoogle Scholar
97. Seeley WW, Crawford RK, Zhou J, Miller BL, Greicius MD. Neurodegenerative diseases target large-scale human brain networks. Neuron 2009;62(1):42–52.
MedlineGoogle Scholar
98. Zhou J, Gennatas ED, Kramer JH, Miller BL, Seeley WW. Predicting regional neurodegeneration from the healthy brain functional connectome. Neuron 2012;73(6):1216–1227.
MedlineGoogle Scholar
99. Prusiner SB. Molecular biology of prion diseases. Science 1991;252(5012):1515–1522.
MedlineGoogle Scholar
100. Liu L, Drouet V, Wu JW, et al. Trans-synaptic spread of tau pathology in vivo. PLoS ONE 2012;7(2):e31302.
MedlineGoogle Scholar
101. Mott RT, Hulette CM. Neuropathology of Alzheimer’s disease. Neuroimaging Clin N Am 2005;15(4):755–765, ix.
MedlineGoogle Scholar
102. Salloway S, Ferris S, Kluger A, et al. Efficacy of donepezil in mild cognitive impairment: a randomized placebo-controlled trial. Neurology 2004;63(4):651–657.
MedlineGoogle Scholar
103. Holland D, Brewer JB, Hagler DJ, Fennema-Notestine C, Dale AM. Subregional neuroanatomical change as a biomarker for Alzheimer’s disease. Proc Natl Acad Sci U S A 2009;106(49):20954–20959.
MedlineGoogle Scholar
104. Ryan R. Update on Alzheimer’s Disease Clinical Trials. National Institute on Aging. https://www.alz.washington.edu/NONMEMBER/SPR12/Ryan.pdf. Published April 2012.
Google Scholar

Article History

Received November 29, 2012; revision requested January 3, 2013; revision received April 22; accepted May 5; final version accepted May 5. From the 2011 RSNA Annual Meeting.
Published online: Dec 2013
Published in print: Dec 2013

Vol. 269, No. 3

Abbreviations

Abbreviations:
AD	Alzheimer disease
ADNI	Alzheimer’s Disease Neuroimaging Initiative
Aβ	amyloid beta
ARIA	amyloid-related imaging abnormalities
CSF	cerebrospinal fluid
FDA	Food and Drug Administration
FDG	fluorodeoxyglucose
MCI	mild cognitive impairment
PIB	Pittsburgh compound B

Metrics

Altmetric Score

PDF download