The latest from Cogstate, leaders in optimizing the measurement of cognition

Research Finds Neurodegeneration In The Absence Of Beta Amyloid Is Not Associated With Cognitive Decline

August 30, 2016

Lancet NeurologyAs we age, our brains inevitably change just as the rest of our bodies do. Once past 60 years old, some of those changes can become noticeable in our daily lives such as small slips of memory. With the increasing information available on Alzheimer’s disease, dementia and other age-related cognitive decline, those signs can cause concern that we may be at the beginning stages of one of those diseases.

Being able to identify the difference between an early disease onset and normal aging is important. Combining diagnostic tools of cognitive testing and brain imaging, medical providers can determine the future path of our current condition.

New research, just published in The Lancet – Neurology, using six years of data from the Australian Imaging, Biomarker and Lifestyle (AIBL) study provides significant findings on the cognitive trajectories of older, healthy adults categorized using a two biomarker framework. These include measurements of brain amyloid β (Aβ) deposition (as either low Aβ1−42 levels in cerebrospinal fluid or high Aβ levels in the brain as measured by positron emission testing) and neurodegeneration (as hippocampal volume using MRI), with neuropsychological testing as a functional metric conducted at baseline, 18 months, 36 months, 54 months and 72 months.

Specifically, the AIBL-Preclinical Alzheimer’s Disease Cognitive Composite (PACC), a variant of the ADCS-PACC, was used to track cognitive performance. Combining tests that assess episodic memory, timed executive function, and global cognition, the ADCS-PACC is the primary outcome measure for the first clinical trial in preclinical AD, the Anti-Amyloid Treatment in Asymptomatic Alzheimer’s study (A4 study).

Previous research has found the PACC to be a sensitive screening tool for early cognitive decline in at-risk but healthy adults. The AIBL-PACC consists of four tests:

  • California Verbal Learning Test (CVLT) (which replaces the Total Recall score from the Free and Cued Selective Reminding Test [FCSRT] found in the ADCS-PACC)
  • Delayed Recall score on the Logical Memory IIa subtest from the Wechsler Memory Scale
  • Digit Symbol Substitution Test score from the Wechsler Adult Intelligence Scale–Revised
  • Mini Mental State Exam total score

As the PACC has evolved, one recent study found that replacing the MMSE with a test of executive function (such as the Z-scores of Attention, Verbal Fluency, and Episodic Memory for Non-demented older adults (ZAVEN) composite) improved the sensitivity of the composite to detect subtle decline.

This two marker method, combining the presence of amyloid beta with neurodegeneration, was recommended by the National Institute on Ageing – Alzheimer’s Association (NIA-AA) as criteria for determining preclinical Alzheimer’s disease. Three stages were defined for disease progression; stage 1 included just the presence of Aβ, stage 2 combined both Aβ deposition plus neurodegeneration, and stage 3 added subtle decline in cognition.

However, a previous study found that two-thirds of healthy, elderly adult participants did not fall into any of these three stages. Specifically, two new groups were recommended; one that had no Aβ deposition present nor any neurodegeneration and a group that had neurodegeneration yet no Aβ in their brain. Labeled as “suspected non-Alzheimer’s disease pathophysiology” (SNAP), they represented 23% of the 450 study volunteers.

Breaking down this recommended framework into four possible categories includes:

  • A+N+ – evidence of Aβ deposition along with neurodegeneration
  • A+N- – evidence of Aβ deposition with no neurodegeneration
  • A-N+ – no Aβ deposition but signs of neurodegeneration (SNAP)
  • A-N- – no signs of either biomarker

While several other studies have found this framework to be useful for classification, tracking individuals over a longer time period would provide insight into the effect of each category on cognitive performance. Some may progress from A-N- to full A+N+ while others may stay in one category.

In the new study, 573 healthy adults, mean age of 73 years, were assessed for Aβ deposition by PET and neurodegeneration, measured as hippocampal volume using MRI. From these two biomarkers, they were categorized into one of the four A-N groups:

  • A+N+ 50 (9%)
  • A+N- 87 (15%)
  • A-N+ 126 (27%) SNAP
  • A-N- 310 (54%)

Tracking this large population over six years allowed the two-factor framework to be tested across an extended timeframe for the first time. As expected, individuals in the A+N+ group showed significant decline across all cognitive testing domains, while the A+N- group showed decline across several areas.

The most interesting finding is that the SNAP group (A-N+) did not have substantial cognitive decline over the six years, very similar to the A-N- group. This suggests that neurodegeneration alone without the presence of Aβ may not put an individual on the path to full Alzheimer’s disease.

In fact, during the study period, only 9% of the SNAP group progressed to either mild cognitive impairment or Alzheimer’s disease compared to 24% of the A+N+ group and 16% of the A+N- group.

“The survival models suggest that A+ is a risk factor for progression to disease with N+ compounding this risk; however, with no significant difference between the A-N+ (SNAP) group and the A-N- (control group), N+ in its own right is not a risk factor for progression,” conclude the study authors. “SNAP is a separate pathophysiological disorder or group of disorders with cognitive and clinical trajectories that are different from the Alzheimer’s disease pathway.”

This confirms previous research but now with a larger population across a significantly longer timeframe.  Being able to observe any disease progression using this two biomarker system with the PACC cognitive summary allows researchers to isolate neurodegeneration alone as a non-factor in the Alzheimer’s disease pathway.

“Despite significantly lower performance than A–N– controls at baseline in some cognitive domains (AIBL, PACC and MMSE), the SNAP group showed no decline in cognition over the 6 years after classification. This pattern aligns with previous findings in which the SNAP groups were shown to progress similarly to their A–N– counterparts on measures of memory, executive function, language, attention, global cognition, MMSE, and CDR.”

“By contrast, the A+N– group exhibited significant decline on MMSE and two episodic memory-driven measures (AIBLPACC and CVLT-II LDFR) and the A+N+ group exhibited significant decline on each of the cognitive measures, again consistent with previous findings.”

Questions or comments?  Please contact Dan Peterson