Scientists who study Alzheimer’s disease have mostly ignored the role of seizures, but that is beginning to change, and new research suggests they may provide insight into the progression of the disease and pave the way for treatments.
It’s no surprise to neurologists that some people experience convulsive seizures in the later stages of the disease. In fact, the second patient ever to receive an Alzheimer’s diagnosis more than a century ago suffered from them. But because brain damage can cause seizures, they were long thought to be just one more casualty of a deteriorating brain.
Now evidence is accumulating that such abnormal electrical activity is far more common and occurs much earlier—and perhaps even precedes obvious signs of memory loss. This raises the possibility that seizures may be intimately tied up with the progression of the disease.
New research that lends credence to this hypothesis was shared at the Alzheimer’s Association International Conference in Los Angeles this week. One study looked at 55 patients between the ages of 50 and 69 who were admitted to an Israeli medical center with their first known seizure. A quarter of them went on to develop dementia—with a mean time to the diagnosis of eight and a half years. Another study of nearly 300,000 U.S. veterans over the age of 55 found that seizures were associated with twice the risk for developing dementia between one and nine years later.
It’s been notoriously difficult to tease out the causes of Alzheimer’s, which is complex, progresses slowly, and may have many different triggers. Finding an association between seizures and dementia is by no means proof of causality, but the new studies excluded people with known risk factors for both epilepsy and Alzheimer’s, such as cerebrovascular disease and traumatic brain injury.
“We had to design a very conservative study that takes these participants out, said Ophir Keret, a neurologist at the University of California, San Francisco who was a researcher on both studies, “because we wanted to isolate the effect that epilepsy would have.”
Michela Gallagher, a professor of psychology and neuroscience at Johns Hopkins University, said the researchers used “good criteria for defining unprovoked seizures independent of things that could be causing [epileptic activity],” and pointed to the large sample size and long time period for tracking each veteran’s health trajectory as two of the study’s strengths. One weakness she noted is that the Veterans Affairs health data used in the study did not allow researchers to differentiate between Alzheimer’s or other types of dementia. Neither the veterans study or the smaller one has been published yet in a peer-reviewed journal.
Two additional studies presented during a poster session at the conference found that people with Alzheimer’s had more seizures than normal controls. However, one of those studies, which is also unpublished, looked for evidence of seizures before a diagnosis of Alzheimer’s, and failed to find such an association. The lead researcher of that work, neurologist Jonathan Vöglein of the German Center for Neurodegenerative Diseases and LMU Munich, nevertheless called Keret’s research “very, very interesting,” and said the results could be in keeping with previous research showing that seizures can occur before the onset of dementia in some people with Alzheimer’s disease.
Lennart Mucke, director of the Gladstone Institute of Neurological Disease and a professor at UC San Francisco, has been examining the relationship between epilepsy and Alzheimer’s in both animal models and people since the early 2000s. He noted in an interview that one reason study results may differ is that some types of seizures are easy to miss. “When we looked at patients who had come to UCSF with both epilepsy and Alzheimer’s disease, it became clear that a lot of the epilepsy they had was non-convulsive,” said Mucke.
Patients with these so-called partial seizures might stop and stare, or experience psychic phenomena like deja vu. Seizures that can be detected on EEG during an overnight testing session are often missed during standard daytime tests that may be as short as 20 minutes.
In a 2016 paper, Mucke, who was not involved in the research presented this week, reported that more than 40 percent of the patients with Alzheimer’s he studied had epileptic activity. Mucke said that he and his colleagues also found in an earlier studythat “the onset of the seizures wasn’t late in the game but could happen shortly before the cognitive decline became manifest.”
Research with mouse models supports the hypothesis that seizures or other epileptic events may sometimes be an early feature of Alzheimer’s disease. Scientists have recorded such abnormal electrical activity in the brains of mice before they went on to accumulate amyloid plaques or tau tangles, both hallmarks of Alzheimer’s. Many of these events had no outward manifestation, suggesting again that the prevalence of epileptic activity may be going unrecognized and therefore underestimated.
Seizures are an extreme example of an imbalance in brain function. Normally, a class of cells called inhibitory neurons, which have received scant attention until recently, act much like the bouncers at a night club. Their job is to manage the timing and flow of brain signals and keep excitatory neurons under control. As we age, inhibitory neurons appear to become less effective, resulting in chronic hyperactivity. Low-level hyperactivity has been detected in the hippocampus—a brain region critical to memory—of both rodents and older people with mild cognitive impairment.
Researchers once assumed that hyperactivity was a compensatory mechanism to make up for a brain that couldn’t keep up with cognitive demands. Now many believe the reverse is true—that hyperactivity is pathological and interferes with memory—and that this imbalance is a core feature of Alzheimer’s disease.
“I think there is a recognition it’s not just seizures which are a real abnormality in the brain,” said Gallagher, “but greater excitability in the earliest stages.”
Gallagher has been putting this theory to the test in clinical trials. A short phase 2 trial completed in 2017 used a common anti-seizure drug called levetiracetam and measured reduced activity in the hippocampus of test subjects coupled with improved performance on memory tests.
Gallagher started a company, AgeneBio, that is now recruiting 830 subjects for a phase 3 clinical trial using a specially formulated low-dose version of the drug. Patients with the type of mild cognitive impairment that precedes Alzheimer’s will take the drug for a year and a half to “crank down the neural activity,” said Gallagher, and see whether the disease progresses more slowly. Three smaller phase 2 trials using a similar strategy are also underway, sponsored by the Medical College of Wisconsin, University of Minnesota, University of San Francisco, and Oxford University Hospitals.
Mucke, like Gallagher and many other neuroscientists, believes that the phenomenon of hyperactivity “is ripe for the discovery of additional therapeutic entry points that might not only be symptomatically beneficial but also have the potential to be disease-modifying.” He points out that brain rhythms have a strong impact on immune system function in the brain, which is increasingly implicated in the pathogenesis of Alzheimer’s disease.
For now, the question remains whether seizures or low-level hyperactivity are a result of Alzheimer’s, an accelerator, an early warning sign, a risk factor, a trigger, or some complex combination. To try to figure that out, researchers point to the need for more trials using sensitive biomarkers to diagnose both seizures and Alzheimer’s, large numbers of test subjects and controls, and long follow-up times.
At a minimum, knowing whether new seizures in older adults are a harbinger of the disease would be valuable. Keret, a fellow at the Global Brain Health Institute at UCSF, points out that even though there are still no proven drug therapies for Alzheimer’s, some lifestyle changes such as exercise or managing known risk factors for dementia may be helpful for delaying the onset of the disease.
So, “if we do establish that this is an early sign, one thing that I would be very interested in evaluating,” said Keret, is whether “interventions at this stage are successful.”