Cognitive Tests, Biomarkers Can Pick Up Alzheimer's, but Treatments Still Don't Offer Much Relief

By Will Boggs MD

April 29, 2020

NEW YORK (Reuters Health) - Brief cognitive tests and biomarkers can identify some patients with Alzheimer dementia, but current treatments provide little relief from symptoms, according to three new reviews.

"The screening tests provide little to no discrimination in the space they are most needed, between mild cognitive impairment and mild dementia. The biomarkers are good but not clinically actionable and have little utility outside of research settings. The medications offer some, albeit small, benefit for cognition," said Dr. David A. Bennett of Rush University, in Chicago, who co-authored an editorial on the research.

"Much more work remains to be done to improve the health and well-being of our aging population," he told Reuters Health by email.

The reviews, published in the Annals of Internal Medicine, evaluated evidence on the accuracy of brief cognitive tests for clinical Alzheimer-type dementia (CATD), biomarker accuracy for classifying Alzheimer disease (AD) in older adults with dementia, and the effects of prescription drugs and supplements for treatment of CATD, respectively.

Based on 57 studies of brief cognitive tests, researchers found the Brief Alzheimer Screen and Montreal Cognitive Assessment distinguished between CATD and normal cognition with median sensitivity and specificity values above 90%, whereas the Mini-Mental State Examination, Memory Impairment Screen, and Test Your Memory yielded sensitivity and specificity values above 85%.

Several memory and language tests also accurately classified CATD versus normal cognition.

In all cases, classification rates were lower for distinguishing CATD from mild cognitive impairment (MCI) than for distinguishing between CATD and normal cognition.

"For the more clinically challenging and relevant questions of distinguishing Alzheimer's dementia from mild cognitive impairment or of distinguishing mild Alzheimer's dementia from normal cognition, the brief tests were only moderately accurate," said Dr. Howard A. Fink of Minneapolis VA Health Care System, who worked on all three reviews.

"These findings suggest that these tests are not adequate by themselves for making a clinical diagnosis," he told Reuters Health by email. "However, they may be useful for identifying who warrants further evaluation or, in patients with a typical course of worsening cognition and function, brief testing may provide sufficient objective evidence of cognitive impairment to support an Alzheimer's dementia diagnosis."

The review of 15 brain imaging studies and nine cerebrospinal fluid (CSF) studies found that amyloid PET and fluorodeoxyglucose (FDG) PET could accurately distinguish AD from non-AD.

MRI accurately distinguished AD from non-AD dementia but less accurately differentiated AD from Lewy-body pathology.

CSF biomarkers beta-amyloid 42 and total tau were moderately sensitive and specific for distinguishing neuropathologically confirmed AD from non-AD, AD from Lewy-body dementia, and AD from frontotemporal-lobe degeneration. But the ratios of beta-amyloid 42 to phosphorylated tau and total tau to beta-amyloid 42 and phosphorylated tau appeared more accurate.

"I think the results may have limited applicability to current clinical practice," Dr. Fink said. "This is due to the methods used in the studies probably not being feasible in typical clinical settings, potential patient selection bias in the studies, and the lack of disease-modifying treatments that might be differentially targeted based on biomarker results."

Finally, the review of treatments included 55 studies reporting cognition, function and global measures of dementia and 12 studies reporting behavioral and psychological symptoms of dementia.

Cholinesterase inhibitors and memantine slightly reduced short-term cognitive decline, and cholinesterase inhibitors slightly reduced reported functional decline, but the benefits over placebo were of uncertain clinical importance.

Evidence was generally insufficient to demonstrate the effects of drug treatment on behavioral and psychological symptoms of dementia.

Similarly, evidence for the effects of supplements on cognition and function was mostly insufficient.

"Though there are many Internet and other claims for Alzheimer's treatment benefits of various supplements, and many state medical marijuana programs include Alzheimer's disease as a qualifying indication, the evidence to support these claims is extremely weak," Dr. Fink said.

"Effects of available prescription drugs on cognition, function, and global outcomes are symptomatic (i.e., they don't change the underlying disease course), small, and of uncertain clinical importance," he said. "Evidence is insufficient about whether prescription drugs are of any benefit on behavioral and psychological symptoms of dementia. Evidence is insufficient about whether supplements are of any benefit on any clinical outcomes in patients with Alzheimer's dementia."

"I think the results should help inform prescribers' conversations with their patients with Alzheimer's dementia and their caregivers about the likelihood the patient will experience clinically meaningful treatment benefit beyond what can be achieved with placebo (i.e., not very high)," Dr. Fink said. "They also should increase awareness that the evidence any supplements or cannabinoids are helpful for treatment of Alzheimer's dementia is very weak (insufficient) or absent."

SOURCE:,, and Annals of Internal Medicine, April 28, 2020.