Biomarker Identification Accelerates Alzheimer Research

Susan Landau, Ph.D. UC Berkeley

Michael Ewers, Ph.D. Trinity College, Dublin

A new study, showing that PET scans and cognitive testing can help detect the risk of developing Alzheimer disease (AD), is among promising new research that could lead to diagnosis of AD at the preclinical stage.

The findings were among those presented by investigators from the landmark Alzheimer’s Disease Neuroimaging Initiative (ADNI) at the Alzheimer’s Association’s 2009 International Conference on Alzheimer’s Disease (ICAD) held in July in Vienna, Austria.

“Aggressive imaging research is under way to identify patients who are at risk for developing or are in the early stages of AD,” said Matthew T. Walker, M.D., chair of the RSNA Education Committee’s Neuroradiology Subcommittee. “One goal is early intervention to slow the progression of disease. The ultimate goal is to identify high-risk patients and prevent the disease from developing. To that end, great strides have been made with fluorodeoxyglucose (FDG) PET and MR morphometry in combination with neurocognitive tests and other measurable biomarkers.”

Launched in 2004, ADNI is an ongoing $60 million public-private partnership organized by the National Institutes of Health (NIH) to test whether imaging technologies such as MR imaging, PET, biomarkers and clinical and neuropsychological assessment can be combined to measure progression toward AD. The private partners are managed through the Foundation for NIH.

This multicenter initiative involves 57 centers in Canada and the U.S. and includes more than 800 people who have normal cognition, mild cognitive impairment (MCI) or the early stages of AD. The initiative is unique because any qualified researcher can access its database at www.loni.ucla.edu/ADNI.

PET, Memory Scans Predict AD

In one ADNI study, investigators from the University of California (UC) Berkeley, discovered that subjects with MCI who had a low baseline FDG-PET and poor memory recall were 15 times more likely to develop AD over a two-year period compared to patients who had normal PET scans and memory recall.

Primary investigator William Jagust, M.D., and Susan Landau, Ph.D., used data from 85 ADNI participants with MCI that included MR imaging, PET, cerebrospinal fluid protein measurements, the genetic marker apolipoprotein E and memory recall tests at six-month intervals. The goal of the study was to use a variety of predictor variables obtained at baseline to identify MCI patients likely to experience further cognitive decline or convert to AD.

Researchers found that low measurements of glucose metabolism in FDG PET scans and poor recall on an auditory-verbal memory recall test were the most consistent predictors for progressing from MCI to AD. Of the total group, 28 subjects converted to AD within the two-year follow-up period, said Dr. Landau, a post-doctoral fellow at UC Berkeley’s Helen Wills Neuroscience Institute and the Lawrence Berkeley National Laboratory.

This is the first time a longitudinal study has examined all of these biomarkers in the same subjects, which aids researchers in comparing the predictive value of any one more biomarker over the other, said Dr. Landau. The research has been submitted for publication, she said.

“This research is important because it will help us select participants for future studies,” said Dr. Landau. “We need to figure out who is more likely to experience clinical decline so we can target those patients for trials and treatments.”

Hippocampus Key to Early AD Diagnosis

Researchers at the University of California (UC), Berkeley, discovered that subjects with mild cognitive impairment (MCI) who had a low baseline FDG-PET and poor memory recall were 15 times more likely to develop Alzheimers disease (AD) over a two-year period compared to patients who had normal PET scans and memory recall. Above: FDG-PET images show reduced glucose metabolism in temporal and parietal regions in AD and MCI.

MR imaging-based measures of brain atrophy in the hippocampus proved to be the most sensitive to early indicators of AD when combined with another primary biomarker or neuropsychological measure as demonstrated by Michael Ewers, Ph.D., and principal investigator Harald Hampel, M.D., both of Trinity College of Dublin. Dr. Hampel presented ADNI findings at ICAD.

In a study of 345 subjects including 81 AD patients, 163 amnestic MCI subjects and 101 elderly healthy controls, researchers used a relatively simple prediction model combining hippocampal volume measured by MR imaging and episodic memory testing to diagnose AD at a very early stage with 94 percent accuracy, according to Dr. Ewers.

“Results show that the fully automated MR imaging-based volumetry of the hippocampus can achieve clinically relevant diagnostic accuracy when combined with psychometric tests of episodic memory ability or cerebrospinal fluid markers of tau and beta-amyloid,” said Dr. Ewers, , senior research fellow at Trinity College.

The research, also conducted by Cathal Walsh, Ph.D., and other ADNI researchers, has been submitted for publication, according to Dr. Ewers, who is analyzing a follow-up study using ADNI data.

“We have demonstrated that a combination of primary biomarker candidates significantly improves early detection of AD when compared to unidimensional prediction of AD,” said Dr. Ewers. “Eventually we will have to weigh the benefits against the cost of the assessment.”

ADNI Database Promotes Information Sharing

Qualified physicians seeking such landmark research can access the ADNI database which contains biomarker data along with thousands of MR imaging and PET brain images and clinical data, according to Neil S. Buckholtz, Ph.D., chief of the Dementias of Aging Branch of the Division of Neuroscience at the National Institute on Aging and a founder of the ADNI initiative. Medical researchers at universities and those who work for imaging and pharmaceutical companies are given equal access, he said.

The database has fast become a model for information sharing—critical considering that approximately 35 million people worldwide are living with AD or some form of dementia, with that number expected to nearly double every 20 years to 65.7 million by 2030, according to the 2009 World Alzheimer Report.

ADNI received an NIH grant utilizing American Recovery and Reinvestment Act funds that will allow recruiting 200 new subjects with an earlier stage of MCI and tracking normal cognitive aging subjects as well as those with a later stage of MCI from the original ADNI study.

“No other study has been able to do this and get as many participants at 57 sites in the U.S. and Canada,” said Dr. Buckholtz.

Dr. Landau said she believes ADNI studies will lead to many breakthroughs in the fight against AD in years to come. “This is an amazing joining of forces of all these researchers,” she said.