The Decomposition of Alzheimer's Disease

The biochemistry of Alzheimer's disease is complex and varied, still incompletely mapped at the detail level. At the edges it merges into grey areas shared with other forms of neurodegeneration - a large number of Alzheimer's patients are diagnosed with other forms of dementia or cognitive impairment. That Alzheimer's is one item in the official list of diseases, that the borders between various forms of neurodegeneration are drawn as they are, is a historical accident carried across more than a century of the taxonomy of disease, not a reflection of current opinions. The age-related dysfunction of the brain is driven by numerous pathological processes. Differences of relative degree between these progresses, and in the locations in the brain that are worst affected, mix and match to produce the various named age-related conditions, collections of different symptoms. The classification of those symptoms into the buckets called diseases happened in most cases long before modern investigations of neural biochemistry. So we have the country of Alzheimer's disease, whose borders as drawn by symptoms and present fairly crude methods of diagnosis encompass what will probably come to be understood as several distinct conditions. They also likely enclose portions of other known conditions, such as vascular dementia, and this muddies the waters in many ways.

In the years ahead, as the first therapies arrive to effectively address the underlying processes that produce neurodegeneration, there will be a redrawing of borders in the matter of brain aging. Some named conditions will vanish, others will split into categories, and entirely new named diseases will arise. In this way taxonomy loosely reflects progress. Being able to remove one cause in a condition that has several causes is one of the most effective ways to figure out how everything fits together, and what the true classification should look like. For Alzheimer's, this phase of research and development is almost upon us. The condition is characterized by harmful accumulations of amyloid-β and tau, to different degrees in different patients, and in different parts of the brain. It is both an amyloidosis and a tauopathy, but without removing one or the other, it is hard to determine the relative importance of these forms of metabolic waste. Even if both are dealt with, there is still the matter of other conditions such as vascular dementia: if a therapy produces little improvement, is it because the target isn't causing significant pathology, or is it because other, untargeted processes are also causing significant pathology? To complicate matters further, the halo of biochemistry surrounding both amyloid and tau aggregates varies considerably by location within the brain and by the form of the aggregate - not all amyloid and not all tau is the same. They are categories, not single items.

Still, not so long ago, researchers finally demonstrated clearance of amyloid-β in the human brain, and in a way that appears to result in decreased symptoms of cognitive decline. Tau should follow in the years ahead. Over the next few years, the understanding of Alzheimer's will greatly increase, as the fastest way to pin down the roles and relative importance of the contributing processes is exactly this, to remove them and see what happens. Beyond the gains in understanding, it has the added bonus of being the most plausible road towards effective therapies, those that can do more than merely gently slow the progression of neurodegeneration. Exciting times lie ahead, and there will be many more papers like this one, in which the existing borders between diseases are questioned in light of new knowledge:

Primary age-related tauopathy and the amyloid cascade hypothesis: the exception that proves the rule?

Extensive data supports the amyloid cascade hypothesis, which states that Alzheimer's disease (AD) stems from neurotoxic forms of the amyloid-beta (Aβ) peptide. Applying the framework provided by the amyloid cascade hypothesis to diagnosing and treating AD has proven problematic. Early neuropathological criteria for diagnosing AD focused on Aβ burden, but this strategy was not optimal given that total Aβ plaques correlate poorly with cognitive impairment and neuronal loss. Several large phase III clinical trials of therapeutics targeting Aβ have failed due to lack of efficacy, prompting reflection as to whether the amyloid cascade hypothesis is invalid. The reason for these failures remain unclear, but some investigators have cited these failed trials as evidence refuting the amyloid cascade hypothesis. Other investigators and pharmaceutical companies have concluded that the design of the trials, which failed to confirm target engagement, were the reason. Another possibility is that Aβ triggers a complex neurodegenerative cascade with a late amyloid-independent phase. The future success of an Aβ-targeting agent is required for final validation of the amyloid cascade hypothesis.

While the heterogeneity of dementing illnesses has complicated efforts to understand the relationship between Aβ and cognitive failure, recent progress in understanding non-AD dementias has put AD into sharper focus. Some of pathologies are more readily differentiated from AD neuropathologically, such as vascular dementia, but this can be difficult to quantify. The TDP-43 proteinopathies (e.g. amyotrophic lateral sclerosis) are largely devoid of Aβ and tau pathology. The more closely overlapping "plaque-only dementia" cases were found to largely represent an α-synucleinopathy (i.e., diffuse Lewy body disease). Another pattern of degeneration, however, which has been variably called tangle-only dementia (TOD), neurofibrillary tangle predominant senile dementia, tangle-dominant dementia, among many other monikers, has received far less attention. Large dementia autopsy series designed to advance our understanding of AD have allowed TOD to come into sharper focus and culminated in the development of a new diagnostic category termed primary age-related tauopathy (PART). New consensus criteria place TOD on a continuum with age-related tangles, that are universally observed in aged brains. Considerable evidence indicates that subjects with PART have a distinct constellation of features that sets them apart from classical "plaque and tangle" AD and other tauopathies. Studying these differences may provide clues to the pathogenesis of tauopathies and refine the amyloid cascade hypothesis.

The neurofibrillary tangles (NFT) in PART are essentially identical to those observed in AD. They are composed of similar tau isoforms, form paired-helical filaments, and are concentrated within neurons. The NFT in PART are localized to the medial temporal lobe. NFT in this distribution can be observed in subjects with normal cognition, mild cognitive impairment and dementia. In cognitively normal elderly subjects, autopsy studies have demonstrated that medial temporal lobe NFT are essentially universal and in a more limited distribution in many younger individuals. In demented subjects, approximately 2-10% of subjects display such tangles without significant amyloid deposition. The proportion of subjects with age-associated memory impairment or mild-cognitive impairment in association with PART might be high. Finally, given that Aβ-deposition is commonly encountered in cognitively normal subjects, "benign Aβ" deposits might be masking an underlying tauopathy in some patients leading to reduced prevalence estimates. Methods for differentiating PART tangles and AD tangles (e.g., biochemical or immunohistochemical markers) would be extremely helpful for answering this question. Tangle-only dementia (TOD) was first described in a series of patients with clinical features that were very similar to those of classical AD. While this category likely included some subjects with other dementing tauopathies, a large proportion have PART as a primary pathological dementing process.

What exactly PART represents has been the matter of debate, with various investigators considering it an AD variant, a frontotemporal dementia variant, or normal (or "pathological") aging. Toxins and infectious causes are also possible, but less likely. Currently, the evidence fails to support a role for Aβ toxicity in PART. Subjects with PART have no Aβ deposition. The possibility that PART is a form of pathological brain aging deserves attention. Mechanical injury in the form of mild yet repetitive traumatic brain injury (TBI) is an established trigger for tauopathy in chronic traumatic encephalopathy (CTE) in elite athletes and boxers. While subjects develop PART in the absence of documented TBI, the hypothesis that these tangles are caused by very mild repetitive "wear and tear" type injury can be supported by three arguments. First, the geometry of the human central nervous system is such that foci of mechanical stress concentration are predicted to include the medial temporal lobe and basal forebrain. Second, the presence of an uncal notch in the medial temporal lobe that overlies the transentorhinal cortex is very common even in the absence of cerebral edema, providing direct physical evidence that this site is a focus of stress concentration. Third, patients with known repetitive mechanical brain injury (i.e., CTE) develop tangles in an overlapping distribution, however more widespread and of greater magnitude. Thus, it is reasonable to hypothesize that the cause of PART is a very mild repetitive mechanical "wear and tear" type of age-related injury.


Hyponatremia can also cause mechanical injury to the brain with repeated osmotic swelling. With age the body has less ability to hold on to sodium and swelling could become an issue in amyloid plaque formation.

Posted by: Greg Marlow at November 22nd, 2016 11:09 AM

Post a comment; thoughtful, considered opinions are valued. New comments can be edited for a few minutes following submission. Comments incorporating ad hominem attacks, advertising, and other forms of inappropriate behavior are likely to be deleted.

Note that there is a comment feed for those who like to keep up with conversations.