In Alzheimer's Disease, Researchers Look Beyond Amyloid

Christine Bahls


July 23, 2019

There is no denying that efforts to solve the Alzheimer's therapy riddle have been nothing short of disastrous: billions of dollars spent and decades of time expended with few therapies to show for it.

Hopes raised, then leveled; trial failure after trial failure.

Many of these trials targeted amyloid, a protein that congregates into plaques in the dementia-afflicted brain. Somehow, the idea that amyloid buildup was causing Alzheimer's disease by kickstarting a toxic molecular cascade swayed pharma companies for decades.[1]

But some researchers in the early 2000s turned their attentions upstream. Knowing that effective therapies would be years off, they began searching for biomarkers they believed would appear prior to the manifestation of Alzheimer's clinical symptoms. The ultimate goal, once other markers were found and the appropriate therapies developed, would be to halt the disease from progressing further.

Multiple biomarkers with varied potential uses have been explored: some to identify disease or disease risk very early in the pathologic cascade, others to show whether therapy[2] is working.

Researchers exploring new biologic signposts like metabolites, blood-brain barrier dysfunction, and retinal changes aren't suggesting to get rid of the existing gold standards — namely an amyloid subtype called amyloid beta-42, as well as levels of another protein called tau. Their goal instead is to find adjunctive screening tools, for reasons both humane and pragmatic: 1 out of 9 people past 65 years of age has Alzheimer's, and disease prevalence is slated to grow exponentially in the coming years.[3]

If the past 20 or so years has exposed anything about Alzheimer's disease, it's the heterogeneous nature of its occurrence. It can show up in someone's fourth decade or seventh. It can share symptoms and genetic mutations with other dementias and neurodegenerative diseases, making diagnoses and treatment development a challenge.[4,5]

We can remove amyloid and there is no change in cognition.

"I had a sense it would be a difficult problem to resolve," says Harvard Alzheimer's researcher Dennis Selkoe, MD, who has been researching the disease since the mid-1970s. "But I was naive enough to think, like every other young researcher, that within 15-20 years we would have a better handle on it."

The biomarker quest has been an exercise in patience. Every researcher interviewed for this story has shown that they are in it for the long haul. "What are the underlying molecular events? That to me is the largest challenge in modern science," says radiologist Clifford Jack, MD, of the Mayo Clinic, who has been studying Alzheimer's for 30 years. "Yes, I'm pretty consumed with it," adds newcomer Sharon Fekrat, MD, a Duke University ophthalmologist who happened into Alzheimer's research a few years ago after noticing differences in the retinas of identical twins. One twin, with diagnosed Alzheimer's, had a severe loss of blood vessels in her retina compared with her unafflicted sibling.

These scientists, most on the biomarker bandwagon for many years, believe that once Alzheimer's symptoms show themselves, the disease becomes refractory. The scores of failed therapeutic studies suggest that they may be right. "We can remove amyloid and there is no change in cognition," says Alzheimer's researcher Howard Federoff, MD, a neurology professor at the University of California, Irvine.

Selkoe agrees. "It's like treating a heart attack with Lipitor—you can't restore neurologic functions once people are forgetful."

The consensus among the biomarker hunters is that no one therapy will stop Alzheimer's. It won't be one lab or one researcher.

The complex pathology at work in Alzheimer's dementia is a neuroinflammatory maze, perhaps reflecting why it's been so difficult to develop effective treatments. As the authors of a recent commentary in Nature Reviews Neurology[6] wrote, "Emerging evidence [has found the existence of] axonal disintegration, synaptic dysfunction and degeneration, innate immune response and neuroinflammation, vascular and cell membrane dysregulation, and brain metabolic dysfunction." They also point out that in addition to amyloid, the brains of people with Alzheimer's often also have accumulations of other proteins, like TDP-43 and alpha-synuclein.

Temporally speaking, a lot of time elapses between conversion from healthy brain to that affected by preclinical Alzheimer's disease, or mild cognitive impairment, as it's called. In the 2012 DIAN-TU study,[7] researchers wrote that even the sporadic type (late onset) disease process probably begins more than 20 years before the clinical onset of dementia. So the upstream researchers may have years to explore at-risk patients.

The consensus among the biomarker hunters is that no one therapy will stop Alzheimer's. It won't be one lab or one researcher. It’s likely to be a joint effort.

"We look across the expanse of disciplines, from clinical imaging to proteomics to biomarkers and genetics," said neuropathologist John Q. Trojanowski, MD, PhD, of the Perelman School of Medicine at the University of Pennsylvania. "The siloed view of [the disease] has nearly disappeared."

Uncovering established Alzheimer's biomarkers requires expensive imaging — imaging not covered by insurance companies. The need for markers of all kinds — those associated with progression, regression, diagnosis, and even comorbidities — is crucial, says Trojanowski, particularly if they're more accessible to patients.

"The most validated and predictive markers are those that examine amyloid and tau in spinal fluids or images of tau or amyloid via PET scan, which [cost] 4500 bucks and have radiation exposure," says John Morris, MD, a neurologist at the University of Washington in St. Louis. "These are very powerful biomarkers. But if I wanted to go out into the community to see who is at risk...we need [easier] screening biomarkers. That's where the retina or blood tests [come in]."

Researchers in Federoff's lab[8] have found that blood levels of a number of metabolites are reduced in people with Alzheimer's. These include histamine, asparagine, aspartate, and citrulline, all involved in metabolic pathways that regulate inflammation, thought to be a key pathologic component in the disease. The same results were seen in those with preclinical dementia and mild cognitive impairment.

For Federoff, the biomarker quest centers around how mitochondrial dysfunction and the resulting inflammation contribute to Alzheimer's.[9] Inflammation is seen in postmortem brain samples from Alzheimer's patients,[10] but determining just how it starts has remained elusive.

It may be that amyloid incites or promotes the formation and spread of tau protein, and that the body's inflammatory response to both protein deposits results in neuronal death. A growing body of research suggests that the initial domino in the Alzheimer's cascade may be an earlier-life infection; amyloid is antimicrobial, after all, and its activity may initially ramp up to help clear, say, a virus, only later to impart collateral damage on neurons of the brain.[11,12]

Another theory links the inflammatory response seen in Alzheimer's to a leaky blood-brain barrier.[13]

Researchers like Harvard neuroscientist Rudolph Tanzi believe that there may be multiple pathologic paths to Alzheimer's, all leading to a revved-up immune system and destructive inflammation. And the sheer number of potential biomarkers being looked at supports his view.

Our "stress hormone," cortisol, is one candidate being explored,[14] as are more advanced retinal screenings and a proinflammatory molecule called chitinase-3-like protein 1, or YKL-40. A recent study found that YKL-40 is present in the cerebrospinal fluid 15-19 years prior to symptoms emerging in patients with Alzheimer's.[15] A protein called neurofilament light chain is also being looked at. It's usually found in myelinated axons but also builds up in the cerebrospinal fluid of people with presymptomatic Alzheimer's.[16]

All of these represent potential biomarkers that could be tested years before an at-risk individual develops any symptoms of dementia.

Researchers can't deny that even among the shifting pathologic tides, amyloid does build up in the Alzheimer's brain. It's just, it seems, not the whole story.

Follow Medscape on Facebook, Twitter, Instagram, and YouTube


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.