The 5th Clinical Trials in Alzheimer’s Disease conference, held 29-31 October in Monaco, showcased the field’s pressing need for biomarkers as outcome measures. Several talks focused on electroencephalography (EEG), a somewhat overlooked technique. EEG directly measures the brain’s electrical activity, and therefore synaptic health. Some researchers believe this approach could provide a more immediate and physiological measure of drug response than do fluid biomarkers and brain amyloid imaging. One talk reported that EEG distinguished between control and treatment groups in a six-month trial of the medical food Souvenaid. Another presentation noted that a form of EEG shows promise for the early diagnosis of AD in a multicenter setting. Better-known AD biomarkers made an appearance, also, with a large French prevention trial reporting significant imaging findings among older adults receiving lifestyle interventions.

EEG shares some similarities with functional MRI. Like that technique, it comes in two flavors: resting state and task related. In the task-related variety, researchers give a visual or auditory stimulus to participants and then record the brain’s response, known as an event-related potential (ERP) (see ARF related news story and ARF news story). Whereas ERPs measure cognitive function, resting-state EEG, like fMRI, can be used to map the brain’s connectivity and organization, said Ilse van Straaten at VU University, Amsterdam, the Netherlands.

Van Straaten presented data from the 24-week Souvenir II trial, which compared Souvenaid to placebo in 259 participants with mild AD at 27 centers in Europe. At last year’s CTAD conference, researchers reported that the intervention improved scores on several memory tests (see ARF related news story; Scheltens et al., 2012). The milkshake-like drink is being developed by Nutricia, a unit of the food company Danone. Souvenaid contains nutrients that promote the growth of neurites and dendritic spines, leading the authors to hypothesize that it improves synaptic function. To test whether this might be true, they turned to EEG, van Straaten said.

To map connectivity, the authors first compared data from 21 electrodes placed across the scalp to record from different brain regions. Van Straaten and colleagues quantified the amount of similarity in the signals using a measure called the phase lag index. Functionally coupled brain regions tend to synchronize their activity, so similar signals reveal connections between regions, van Straaten said. The researchers then used the coupling data and modern network theory to determine the organization of the brain. In a healthy brain, each electrode makes many local connections and only a few long-distance connections, because this is the most efficient model, van Straaten said. In AD brains, by contrast, functional coupling declines and the networks become disorganized.

In the Souvenaid trial, participants taking the supplement maintained stable connectivity and organization, while the control group declined on both of these measures, van Straaten reported. “That was really surprising for us, because we did not know that in 24 weeks, AD patients would show deterioration,” she said. The results suggest that EEG has potential as a clinical trial outcome measure and should be investigated further, she added.

Van Straaten noted that few researchers perform this type of complicated analysis on resting-state EEG data. “I think few people know about the possibilities EEG has in this respect,” she said. The method is not new, however. For example, Andrew Leuchter at the University of California, Los Angeles, has used EEG to analyze brain connectivity in depression and Alzheimer’s disease (see Leuchter et al., 2012; Cook and Leuchter, 1996; Leuchter et al., 1992). The methodology has become more accessible in the last couple of years because EEG data moved to a digital format, making sophisticated computer analyses possible, van Straaten said. EEG has the advantage of being cheaper and more readily available than fMRI, but it suffers from a much lower spatial resolution.

Many researchers believe EEG also holds potential for early diagnosis of cognitive problems. Since synaptic function falters early in AD, before amyloid plaques form, quantitative EEG and ERPs may detect the first stages of disease, said Karim Bennys at Montpellier University Hospital, France, in his talk at CTAD.

Greg Jicha at the University of Kentucky, Lexington, told Alzforum that ERPs show promise as a screening tool for widespread clinical use. The problem is that EEG measurements typically require a skilled technician, limiting their use to specialized settings. To make ERPs feasible in clinical practice, the method will need to be easy to use, reproducible, and largely automated, Jicha said. Jicha participates in an ongoing U.S. trial at six clinical sites to evaluate the effectiveness of a handheld ERP system made by biotech company Neuronetrix, Louisville, Kentucky. At CTAD, company president K.C. Fadem reported that among the first 100 participants, those with probable mild to moderate AD had distinct ERP waveforms from age-matched controls, with the difference reflecting poorer memory and cognitive performance in AD patients. Measurements varied little from site to site, Fadem claimed, suggesting the method is reproducible.

ERPs could also make good trial outcome measures, Jicha said. “Pharmaceutical companies want cheaper, less invasive biomarkers, and have shown a lot of interest in this,” he told Alzforum. He noted a caveat, however: ERPs will not necessarily distinguish between symptomatic and disease-modifying effects. “A drug that makes nerve cells fire better might be potent in restoring normal ERP function and yet might not touch the disease state,” Jicha pointed out. “We are always going to have a need for molecular biomarkers.” These would include amyloid imaging and fluid measurements of tau and Aβ, he said.

Lon Schneider at the University of Southern California, Los Angeles, suggested that a positive ERP signal might not always translate to better cognitive performance. For example, α power is associated with attention and memory, but a drug that improves α power will not necessarily pump up these faculties. Nonetheless, he thinks EEG measurements have value in clinical trials. “EEG/ERP provides methods for assessing that a drug is directly or indirectly affecting brain function, and within a narrow time interval of millisecond resolution,” he told Alzforum. “In my view, it is important for Phase 1 and 2 development to be able to demonstrate neurotropic and psychotropic effects of brain-active compounds.” He also noted that EEG would provide a more sensitive marker for drug effects than fluid or brain imaging biomarkers.

Meanwhile, at CTAD, preliminary results from the Multidomain Alzheimer Preventive Trial (MAPT) in France focused on such imaging markers. This ongoing three-year prospective study, funded in part by the French Alzheimer Plan, investigates whether lifestyle interventions can slow cognitive decline or lower the risk of AD (see ARF related news story). More than 1,600 cognitively healthy adults over 70 years of age receive one of four interventions: placebo; omega-3 supplements; a multi-domain intervention consisting of nutrition advice, exercise, cognitive training, and social activities; or both omega-3 plus the multi-domain intervention. The study will conclude in March 2014, with the main outcome measure being performance on the Free and Cued Selective Reminding Test. This study resembles other European initiatives such as the Finnish Geriatric Intervention Study to Prevent Cognitive Impairment and Disability (aka FINGER). At CTAD, several talks presented baseline and preliminary data from three MAPT imaging studies performed on subsets of participants.

In the smallest study, 68 participants received FDG-PET scans, which measure brain glucose metabolism. Half of these people got the multi-domain intervention and half did not; both groups contained a mixture of people on omega-3 supplements and placebo. Those who received the multi-domain intervention showed improved brain metabolism over baseline at six months, and remained stable at 12 months, while the control group declined at both time points, project leader Bruno Vellas of the University of Toulouse, France, told ARF. This hints that the intervention might slow degeneration, and supports FDG-PET as a useful outcome measure, he said.

In a second study, more than 400 participants had a baseline structural MRI scan, which will be followed by a final MRI at the end of the study. In the baseline data, people who were judged frail, meaning they walked slowly or could not perform some activities of daily living, had lower hippocampal volumes than their peers who were not frail. Small hippocampal volume is associated with AD, suggesting that frailty might provide a cheap surrogate for dementia risk, Vellas said. A third study examined baseline amyloid PET scans in about 200 people. Nearly 37 percent of participants had a positive brain amyloid signal, with this group also having lower cognitive scores on the Mini-Mental State Exam, Clinical Dementia Rating, and the Free and Cued Selective Reminding Test than those who were negative for brain amyloid. This amyloid-positive percentage is similar to what is reported in the literature, but a bit lower than that seen in the Alzheimer’s Disease Neuroimaging Initiative, Vellas said. As Mike Weiner of the University of California, San Francisco, noted playfully, this goes to show that drinking red wine and eating duck and foie gras is good for your brain.—Madolyn Bowman Rogers.

This is Part 6 of a seven-part series. See also Part 1, Part 2, Part 3, Part 4, Part 5, and Part 7. Read a PDF of the entire series.


Make a Comment

To make a comment you must login or register.

Comments on this content

No Available Comments


News Citations

  1. EEG: Coming in From the Margins of Alzheimer’s Research?
  2. EEG: Old Method to Lend New Help in AD Drug Development?
  3. Nutrient Formulation Appears to Grease Memory Function
  4. Europe Asks If Reforming Health Habits Can Prevent Dementia
  5. CTAD: Turning the Ship Around Toward Early Trials
  6. CTAD: New Data on Sola, Bapi, Spark Theragnostics Debate
  7. CTAD: Regulatory Science Gains Prominence in AD Research
  8. CTAD: Adaptive Antibody Trial to Try Bayesian Statistics
  9. CTAD: ApoE Carriers Sought for Trial of NSAID Derivative
  10. CTAD: AD Treatment Might Not Lower Healthcare Costs

Paper Citations

  1. . Efficacy of Souvenaid in Mild Alzheimer's Disease: Results from a Randomized, Controlled Trial. J Alzheimers Dis. 2012 Jun 26; PubMed.
  2. . Resting-state quantitative electroencephalography reveals increased neurophysiologic connectivity in depression. PLoS One. 2012;7(2):e32508. PubMed.
  3. . Synaptic dysfunction in Alzheimer's disease: clinical assessment using quantitative EEG. Behav Brain Res. 1996 Jun;78(1):15-23. PubMed.
  4. . Changes in brain functional connectivity in Alzheimer-type and multi-infarct dementia. Brain. 1992 Oct;115 ( Pt 5):1543-61. PubMed.

Other Citations

  1. Read a PDF of the entire series.

External Citations

  1. Souvenir II trial
  2. U.S. trial
  3. Multidomain Alzheimer Preventive Trial (MAPT)
  4. Finnish Geriatric Intervention Study to Prevent Cognitive Impairment and Disability

Further Reading