“We demonstrated that cognitive markers were consistently significant and generally stronger predictors than biomarkers, and moreover, that conversion itself did not so much reflect a shift in the neurobiologic characteristics of the disease but rather a large functional decline,” the study team reports in the September issue of the Archives of General Psychiatry.

In an email to Medscape Medical News, senior author Terry E. Goldberg, PhD, from the Litwin Zucker Alzheimer’s Disease Center, Feinstein Institute for Medical Research in Manhasset, New York, urged caution in interpreting this finding.

“Biomarkers unarguably work,” he said. “However, cognitive markers, which are less expensive and less invasive, also work and provide strong complementary information.”

“In my estimation, work on validating new and more sensitive cognitive markers should proceed apace with work on biomarkers,” he added.

Results “Not Surprising”

Reached for comment, Marilyn Albert, MD, from Johns Hopkins University School of Medicine, Baltimore, Maryland, said, “The results are not that surprising, in the sense that it is challenging to predict who is going to progress among those who are mildly impaired.”

“The biological biomarkers that are available tell you something,” she added, “but they don’t tell you everything, and in the end, how people are doing functionally — what their cognition is, what their functional abilities are — is kind of the final common pathway. They’re the things we use to judge whether or not someone is demented.”

Dr. Albert chaired the MCI working group convened by the National Institute on Aging and the Alzheimer’s Association to update criteria established by the National Institute of Neurological Disorders and Stroke/Alzheimer’s Disease and Related Disorders Association (now the Alzheimer’s Association) in 1984.

The updated criteria were published April 19, 2011, in Alzheimer’s & Dementia: The Journal of the Alzheimer’s Association, and reported by Medscape Medical News at that time.

Dr. Goldberg and colleagues note in their article that several comprehensive reviews have demonstrated the predictive power of various biomarkers, and in particular amyloid beta1-42 and tau, in identifying an individual’s likelihood of progressing to AD.

For example, it is known that cerebrospinal fluid (CSF) levels of amyloid beta1-42 and total tau can discriminate between healthy individuals and those with MCI. Neuroimaging studies have shown that hippocampal atrophy is a key brain volume biomarker and is also a predictor of conversion from MCI to AD. Numerous cognitive markers are also known to have predictive power for conversion to AD.

Dr. Goldberg and colleagues say their analysis is the first that they are aware of to examine combinations of cognitive markers, brain volumetric markers, and CSF biomarkers to predict conversion to AD.

Included in the study were 116 individuals with MCI who converted to AD within a 2-year period, 204 individuals with MCI who did not convert during this time, and 197 individuals with normal cognitive function throughout the study period.

As participants in the Alzheimer’s Disease Neuroimaging Initiative, the participants completed a battery of standard tests for cognitive, neuropsychological, and global functioning. CSF concentrations of amyloid beta1-42 and total tau were determined at baseline and at 12 months. Magnetic resonance imaging was used to assess entorhinal and lateral temporal cortical thicknesses and the hippocampal, ventricular, and whole brain volumes. All patients were genotyped for APOE.

In a systematic series of stepwise logistic regression analyses that included variables from all classes of markers (biomarkers, cognitive markers, and risk factors), 3 baseline variables predicted conversion to AD within 2 years. Two of these measures pertained to delayed verbal memory, and the other was left middle temporal lobe cortical thickness.

Table. Significant Predictors of Conversion During 2 Years of Follow-Up

CI = confidence interval

Of note, the researchers found that a decline on the Functional Assessment Questionnaire and the Trail Making Test, part B, accounted for nearly 50% of the predictive variance in conversion from MCI to AD.

“The APOE status and CSF levels of tau or amyloid beta1-42 did not add predictive value to this composite model,” the authors note.

This study, Dr. Goldberg commented, shows that “cognitive markers, especially those involving memory after a delay and, to a lesser extent, executive function, are robust predictors of MCI to AD conversion.”

Interpret Cautiously

The researchers emphasize in their report that biomarkers may have differential predictive power at different times in disease progression.

“Biomarkers are dynamic,” they write, and will vary in how informative they are depending on when they are measured. “They may be most informative in the very early prodromal stages, a perspective that has been incorporated into proposed diagnostic criteria for preclinical AD,” they note.

“It’s not as if the biological markers aren’t important,” Dr. Albert said, “but you can gain so much more accuracy if you have the cognitive measures and the functional measures in addition.”

“Particularly, the Functional Assessment Questionnaire,” she said. “To begin to see declines on that, then it’s telling you that someone is not able to manage on a day-to-day basis anymore and that’s the heart of dementia.”

This study was supported by grants from the Litwin-Zucker Alzheimer’s Center and by a grant from Instituto de Salud Carlos III. Dr Goldberg reports that he has consulted for Merck and GlaxoSmithKline, receives royalties for use of a cognitive test battery (Brief Assessment of Cognition in Schizophrenia) in clinical trials, and has received an investigator-initiated grant from Eisai/Pfizer.

Arch Gen Psychiatry. 2011;68:961-969. Abstract