Exploring EEG as a Biomarker for Neurodegenerative Disease Progression

Neurophysiologist Stephen Morairty, Ph.D., first got his feet wet in sleep research as an undergraduate performing experiments on his rugby team as part of a thesis project, encouraged by professor who was a sleep researcher himself.

In graduate school Morairty watched a talk by Professor Dennis McGinty, Ph.D., a basic sleep researcher from the University of California, Los Angles, and was so inspired he followed McGinty to UCLA to receive his Ph.D. in neuroscience and work in his lab, and then continued on to do his post doc work at Harvard.

When Morairty, now director of in Vivo Physiology and Pharmacology at SRI International’s Center for Neuroscience, started his lab at SRI 16 years ago with Thomas Kilduff, it was a pretty straight-forward sleep and circadian rhythms lab, he told Bioscience Technology. But new information over the last 15 years that links sleep disruptions to virtually every central nervous system (CNS) disorder, has expanded the scope of his work greatly to involve neurodegenerative or neurodevelopmental diseases such as Alzheimer’s, autism, and Huntington’s disease.

“We’re finding out that the systems involved in sleep regulation appear to be attacked in many of these disorders and sleep disruption becomes a primary symptom of many of these things,” he told Bioscience Technology. In some cases, one of the very first signs of a disorder is a disruption in sleep/wake patterns.

Along with looking at sleep as in indication for disease, another thing that has propelled Moriarty’s work into this sphere is the familiarity with electroencephalogram (EEG) recordings, which is a standard technique used in humans and animals to define the sleep and wakefulness electrophysiology, and has translational potential.

In a recent study, published in February in the journal Sleep, Morairty and colleagues examined EEG in mouse models as a potential highly sensitive biomarker for Huntington’s disease that could predict the onset and progression of the disease before any symptoms began to show.

The need for a translational biomarker

Huntington’s disease, which impacts thinking, movement, and cognitive abilities, is unique among neurodegenerative diseases because it is the only one where there is a known cause. It is a single inherited gene mutation caused by the mutant huntingtin protein that causes the progressive degeneration of nerve cells in the brain. People who have the disease can get a blood test and know from birth if they have the mutation, and based on how many CAG repeats a person has, Morairty said, the earlier symptoms will occur, though it is not exact. Beyond that there is no other tracker for the disease.

“We have no other way of testing if the disease course has changed if we develop a potential therapeutic and give it to people who have Huntington’s, there’s no good marker for testing whether the new treatment is having the desired effect,” Morairty said.

Read more: Potential Huntington’s Treatment Successful in Animals, Moves to Clinical Testing

He explained that this puts people who are developing potential therapies in a difficult place, because it may involve clinical trials that last 10 to 20 years to see if there is an effect.

“It’s a really expensive clinical trial for a small population of people and it’s not very viable,” Morairty said.

Pharmaceutical companies and lab tech companies working on CNS disorders have a difficult time with translational assays from preclinical to clinical models, according to Morairty.

“One of the main reasons why there’s such a dearth of new medications coming out for CNS indications, is because there is very little translational – it’s like playing in the dark a lot of times.”

That is where the EEG comes in. An EEG is an objective electrophysiological measure that is a read out of the network underneath the electrode, or a summation of all of the inhibitory and excitatory currents that are going on. It is collected regularly in humans, is noninvasive and is relatively inexpensive, compared to techniques such as MRI or PET scans.

“The EEG has been collected in basically the same way for 80 years, with a few very small enhancements,” Morairty said. “So we had hope that by using new mathematical techniques to deconstruct it, looking at quantitative EEG analysis that we might be able to see changes in patterns that are similar in animal models and in human models.”

He explained that while the behavior of a rat, monkey, or a human is quite different overall, that the structures of the brain are remarkably similar.

“In gross areas of the brain, like what the hypothalamus does in a rat, is very similar to what it does in a non-human primate, and very similar to what it does in a human,” Morairty said. Similarities are also seen in the function of the cortex, and there is similar gross EEG patterns during active waking animals as active human models. When humans go to sleep their EEG changes in a very characteristic way that is the same in animal models.

So, Morairty and colleagues hypothesized, why wouldn’t there be translational changes with disease progression?

With Huntington’s disease it’s known that there are changes occurring along the way, as the mutant protein builds up over decades of life, and eventually reaches a tipping point where a cell becomes sick and eventually dies off. At a cellular level there are changes that are occurring but there is no way to monitor the activity of a million individual cells in human patients.

Morairty said that the EEG will not give any kind of single cell readout, but the benefit is that it tells you what is going on with the network. So as cells become sick, their electrophysiological properties change, and eventually there will be a change in that network that comes before any noticeable changes in behavior.

Looking through quantitative EEG analytical techniques, Morairty tested his hypothesis in the most popular Huntington disease mouse models and wild-type mice and was very pleased with the results.

“The EEG patterns in these rodent models, the changes preceded any measurable changes in motor, in movement, in cognition, or in sleep/wake patterns,” Morairty said. If translational to humans this could be a big advance.

The team found that although the Huntington’s disease mice and the wild-type mice start off very similar, their EEG began to change in early age, and continued to progress until it was extremely different from the wild-type, even before behavioral changes.

One cautionary note is that when EEG recordings are taken from mice, the electrode is implanted under the skin, unlike on the surface of the scalp for humans. This means there is much better signal to noise ratio on the EEG output, so there are very sensitive areas that scientists are able to detect changes in the rodents. It would still be a work in progress to see how researchers could enhance those signals for humans.

One thing the team hasn’t shown, and what Morairty considers to be especially important, it what would happen to the EEG if the mice were given a treatment to cause the disease to either plateau or reverse.

“The EEG should show that, so if it really is a useful biomarker then we have to show that when you modulate progression of the disease that the EEG represents that.”

There are a few potential new therapies, one developed by Ionis Pharmaceuticals that is meant to knock down levels of the mutant huntingtin protein in the brain with hopes of stopping, or abating progression of the disease. The treatment is in Phase 1 human trials right now, and Morarity has a proposal in with the foundation, and associated with the pharmaceutical company to see if he can get his hands on the treatment to test in the same animal model that he just published on.

If approved, he will examine what happens to the EEG when levels of mutant protein are lowered in mice, which he predicts will change with the course of disease progression. “If it doesn’t then it’s not a very useful biomarker and my hypothesis is wrong,” he said. “But if it does, that shows that the EEG can be used clinically to measure the effectiveness of this new treatment.”

Stay tuned for more from Stephen Morairty in an upcoming article that delves into new technologies being utilized in sleep research, such as optogenetics, DREADs, and miniaturized endomicroscopes.