Ronald C. Petersen, MD, PhD: Alzheimer disease is essentially defined by 2 proteins. One is amyloid, that makes up the plaques—plaque and tangle disease—and the protein tau, which makes up the tangles. So those 2 proteins define the disease. A major line of thinking in the theory of how Alzheimer disease evolves is that amyloid is the initiating process. It comes from this precursor protein called APP, amyloid precursor protein, and that protein then gets cleaved by enzymes later on.
Usually, an enzyme called alpha-secretase cleaves the protein and the harmless residues are eliminated. Occasionally, in some people, instead of alpha-secretase doing its work, the secretases beta secretase and gamma secretase intervene and cleave the protein to leave a residue of maybe 38, 40, 42 amino acids. This is a particularly nasty residue that accumulates in the brain. It is quite sticky and goes down a pathway then to ultimately produce the amyloid neuritic plaque.
So the amyloid cascade hypothesis sort of takes that whole sequence of events into account and suggests that if we intervene at various points in that cascade, we will be able to stop this abnormal processing of amyloid. Of course, this is all still theoretical, but there’s a good deal of animal, genetic, and human data that would support that.
Alireza Atri, MD, PhD: We’re coming to an appreciation that’s a bit different about the pathobiology and the trajectory of the pathobiology of Alzheimer disease and related dementias, or the neurodegenerative conditions. One of the appreciations really is that even though Alzheimer is defined by amyloid plaques and neurofibrillary tangles, that’s a manifestation of disease for which, in the new framework that has been recently released and conceptualized by the NIA [National Institute on Aging] and the Alzheimer’s Association, one can think about amyloid and one’s amyloid pathology stage in the brain. Amyloid is a protein that we don’t understand quite yet, regarding what its function is. It may have something to do with synapses in the brain, and even potentially the learning or signal transduction. But this amyloid protein, the 42 version of it, and the ratio of 42:40 is toxic, comes in different forms. When it’s finally processed by these enzymes, BACE [beta secretase], and we’ll talk about BACE and things like it, you get monomers. And then they develop, they stick together, they become maybe dimers and then multiple oligomers. Then they kind of sit there in protofibrils, develop fibers, so fibrils, ultimately in a plaque. That evolution can take quite a while.
It turns out that some of the earlier forms may be toxic to synapses. So that’s one appreciation—that there’s a very big lag time between people showing symptoms and actually having different stages of amyloidosis in their brain. Somewhere along the pathway, amyloid induces neurofibrillary tangles. So this tau protein, which is inside the cytoskeleton, is very important in axonal transport in the neurons, and it becomes hyperphosphorylated and dysfunctional.
The way I describe it, amyloid seems to be toxic kindling. So it’s sitting around in the brain being toxic to synapses, kind of making them dysfunctional. But really, when you have the fire is when the match is lit, and it’s really the tangles that spread. And tangles, we’ve known for a very long time, correlate with much more neuronal damage and also spread in a relatively well-known pattern. They also go along with people who are having symptoms.
So there’s the amyloid kind of score pathway, there’s tau, and then, of course, there is the effect of amyloid and tau on nerves—so, neurodegeneration. We can potentially measure that in different ways. On MRI [magnetic resonance imaging], we can look at atrophy, for example, but we’re now getting these biomarkers that we’re still developing to see synaptic damage with neurogranin or axonal damage with neurofilament light. And also, there’s the vascular component. The brain is not sitting outside the body. It’s sitting inside the body. Vascular damage accrues. Therefore, inflammation accrues. Inflammation can accrue in multiple ways in these pathways, or by the effect of toxins or pathogens, for example.
It’s a very complex story that makes things very heterogeneous. Both our environments are heterogeneous, and our genetics are heterogeneous. So individuals have the combination of resilience factors, or reserve, and also vulnerabilities. What we’re appreciating is for an individual with late-onset dementia syndromes, cognitive impairment or dementia, chances are if you say Alzheimer pathology is present, you’re right. You’re probably 80%, 90% correct in the right setting. But they may not be the only components. There are amyloid pathways and non-amyloid degenerative pathways with other proteins—TDP-43, for example, tau, Lewy bodies—and then there are age-related pathways. In an older individual, neuropathology studies are showing once you get into your 80s, and definitely into your 90s, you tend to have more than 1 pathology, as opposed to early onset syndromes, where the diseases tend to be purer in individuals who are in their 40s or 50s.
Just very recently, in the last few weeks, there’s been a consensus group with a very important paper that came out in the journal Brain, about the criteria known as LATE. It’s a limbic-predominant age-related TDP-43 proteinopathy that is thought to be present in 15% to 20% of individuals who are age 85 and above, let’s say, that present with a typical Alzheimer-like amnestic syndrome but don’t actually don’t have Alzheimer pathology. So we’re appreciating that not all syndromes, even though they may be very amnestic-like and Alzheimer-like, especially in older individuals, are caused just by Alzheimer disease. The story is much more complex. We have to really be able to parse out the underlying pathobiology and come up with specific diagnosis pathways to come up with the cause and contributing factors. We’re trying to do that with biomarkers and also with guidelines, as far as how to evaluate individuals with symptoms.