Diagnostics & Imaging Week

For nearly four years, Neuronetrix (Louisville, Kentucky) has been developing a test that can accurately diagnose Alzheimer's disease (AD) in patients, as well as other cognitive impairment disorders.

Last week the company sowed the strides it has made, reporting positive preliminary results from its Cognision System at the Society for Neuroscience (Washington) annual meeting in Washington.

The Cognision system, which is Neuronetrix's sole product, uses event-related potentials (ERP) to record brain activity while the patient listens to a sequence of sounds. ERP is the electrical activity produced by the brain in response to a sensory stimulus or associated with the execution of a motor, cognitive, or psychophysiologic task.

Patients with Alzheimer's disease process auditory information differently than healthy individuals, and that difference can be detected with the test.

"What we did was use an oddball paradigm for sound in the test," K.C. Fadem, Neuronmetrix COO, told Diagnostics & Imaging Week. "One sound is played commonly, while the other sound is played less frequently. If the brain can't detect novel stimuli, then it is impaired to learn. In Alzeheimer's disease patients, there are several parts of the brain that are impaired."

The test, however, isn't geared toward detecting if a patient has the potential to suffer from the disease 20 years down the line, according to the company.

"Yes, there have been discussions about testing for genetic diseases in the medical community, but there isn't a market for a product that says 'Yes you will have a disease 20 years in the future,'" Fadem said.

Nearly 22 patients suffering from Alzheimer's participated in the initial clinical study, held at the University of Kentucky's Sanders-Brown Center for Aging (Lexington).

This phase of the study was primarily performed to evaluate the Cognision System in a real-world clinical setting to test ease of use, patient tolerance, and most importantly, data quality.

"Clinical usability and technical performance were major milestones in our study at the Sanders-Brown Center," Fadem said. "The system exceeded our expectations in terms of data quality and we demonstrated the test could be efficiently performed on real Alzheimer's disease patients."

The ability of the ERP method to diagnose Alzheimer's disease has been demonstrated in scientific studies at several research laboratories around the world. In a scientific paper published in 2007, Neuronetrix collaborators Dr. Robi Polikar, et al., reported that their ERP classification accuracy "exceeded that of the trained community clinic physicians, and closely approached the gold standard performance of the university hospital-based clinic evaluation."

Fadem said, "The problem with using ERP in the past is having a user friendly system that can be used by untrained technicians in a clinical setting."

The next clinical study phase for Neuronetrix will be to test the Cognision System in a multi-center trial on patients who have been diagnosed with Alzheimer's disease using a rigorous diagnostic protocol. This protocol will be similar to that used in the Alzheimer's Disease Neuroimaging Initiative (ADNI), a large international study sponsored by the National Institutes of Health.

The ADNI protocol uses psychometric testing, MRI, and cerebrospinal fluid analysis (CSF) to accurately characterize several aspects of Alzheimer's disease. Neuronetrix will use these same tests to correlate with the Cognision ERP results.

"What we'll be doing is having 200 patients at four sites, with three in the U.S. and one in Europe," Fadem said. "In the second study, we will train our system to distinguish Alzheimer's patients from patients (who don't have) cognitive impairment."

Company Chairman Dr. John Barker said, "We want to compare our Cognision System with the most advanced diagnostic methods possible to clearly demonstrate the system's capabilities."

Neuronetrix is working with several NIH Centers of Excellence in the U.S. and a center in Europe to organize the study, which is expected to begin in 1Q09.