Subscribe to eTOC

In the Pipeline-Alzheimer's Disease
Why a PET Scan “Negative” for Amyloid Does Not Rule Out Alzheimer's



CONVERSIONS TO MILD COGNITIVE IMPAIRMENT (MCI), Alzheimers disease (AD), or florbetapir-positive status

Researchers reported that amyloid-beta accumulation in cognitively normal, baseline-negative individuals was associated with poorer trajectories on memory (but not executive function).

A new study designed to detect Alzheimer's disease (AD) at its earliest stages found that some people began to experience subtle memory decline even when brain amyloid-beta (Abeta) levels measured on positron emission tomography (PET) scans were in the “negative” range.

The study, which was published in the March 23 online edition of Neurology, lends support to the widely-held belief that AD is a long process that begins years, and perhaps decades, before it becomes clinically evident. But it adds further insight into the development of AD by suggesting that even slight elevations in brain amyloid within the so-called negative range may be linked to cognitive changes.

The researchers arrived at their findings after giving study participants, all older individuals who were deemed cognitively healthy at the start, periodic PET scans and cognitive tests over a span of about four years.

“These results indicate that even individuals categorized as ‘A-beta negative’ may be in preclinical stages of AD, and that subtle memory decline is an early consequence of cortical amyloid-beta deposition that can be detected within the negative range,” the researchers reported.

The study's lead researcher, Susan M. Landau, PhD, associate research scientist at the Helen Wills Neuroscience Institute at the University of California, Berkeley, said “people in the negative range are thought to be free of Alzheimer's disease pathology and, in general, this is true.” In the cognitively normal people they studied, however, amyloid that was negative on PET scan but gradually increasing coincided with memory decline, she said.

“There is variability within the normal range,” Dr. Landau told Neurology Today. “Even small increases in amyloid are linked to small subtle declines in memory performance.” The changes may not even be noticeable to the individual, she added.

The study findings have no clinical relevance currently since PET scanning to measure Abeta is not typically done in cognitively normal people, but they could help inform researchers designing clinical trials, particularly those involving AD prevention therapies.


LONGITUDINAL FLORBETAPIR trajectories (A) for baseline-negative or baseline-positive cognitively normal individuals who had florbetapir scans at approximately two-year time intervals. For visual clarity, only individuals with >2 florbetapir scans (65% of normal controls) are shown. (B) Mean and standard deviation of annual florbetapir standardized uptake value ratio (SUVR) slopes are shown for baseline florbetapir-negative and florbetapir-positive individuals.

The study authors noted that the “appearance of widespread cortical amyloid-beta occurs in about 25-30 percent of the cognitively normal older population over age 70, and it is hypothesized to be the initial event in the pathological cascade that leads to AD.” Recent research has shown that amyloid-beta positive status is associated with cognitive decline even in cognitively normal people, the study authors said.

“However, the earliest stages of amyloid-beta deposition, before amyloid is widespread throughout the cortex remains poorly understood,” they wrote. “While some studies have reported no relationship between cognitive change and beta amyloid deposition in the negative range these observations could be due to relatively limited follow-up and small sample size.”


To further examine the question, the researchers focused on a group of 142 cognitively normal individuals enrolled in the Alzheimer's Disease Neuroimaging Initiative, a multicenter longitudinal biomarker study. The participants were on average 74.7 years old at baseline (55 to 90 years old was the range) and 45 percent of them were women. They were all considered amyloid negative at baseline and had at least two (18F)-florbetapir PET scans over an average of 3.9 years of follow-up. The goal of the study, which was funded, in part, by the National Institutes of Health and the US Department of Defense, was to determine whether amyloid accumulation was associated with longitudinal changes in memory or executive function.

The cutoff between “negative” and “positive” brain amyloid is based on amyloid pathology detected at autopsy, the gold standard for defining the cutoff, according to Dr. Landau. Commonly used standardized tests were administered to assess cognitive changes.

Over the four to six years of follow up, 13 of the 142 participants (9.2 percent) converted to an amyloid positive status as measured by florbetapir accumulation on PET scan. Fourteen of the 142 (9.9 percent) converted from being cognitively normal to a different diagnosis; 12 of them were diagnosed with mild cognitive impairment (MCI) and two with AD. (One of those who converted to MCI had also converted to an amyloid positive status.)

“Across all baseline negatives, florbetapir accumulation was associated with poorer memory trajectories, but there was no association with executive function,” the researchers reported.

Factors such as apolipoprotein E4 status, age, sex, education, and baseline memory did not seem to influence whether a person would have changes in amyloid. However, the researchers did find that greater accumulation was associated with higher baseline Abeta, indicating that accumulation was ongoing prior to the start of the scanning.

“A key feature of our finding is that slight elevations in amyloid-beta within the negative range have not been previously linked to cognitive changes,” the researchers wrote. But they acknowledged that the study left many questions unanswered.

Dr. Landau said it is important to emphasize that “not all older people are accumulating amyloid. Most of the people we studied — 82 percent — were still amyloid negative and still cognitively normal by the end of the study,” so the big picture is not totally discouraging. “Many people never cross that threshold.”

Dr. Landau said her group is now interested in investigating the involvement of tau, another biomarker for AD, in people whose PET scans are in the negative range.

She said understanding the very earliest stage of AD is important for both drug development and prevention efforts because “once enough pathological change and cognitive change is underway it may not be reversible.”


DR. SUSAN M. LANDAU: “There is variability within the normal range. Even small increases in amyloid are linked to small subtle declines in memory performance.”


David S. Knopman, MD, FAAN, professor of neurology at the Mayo Clinic in Rochester, MN, told Neurology Today that the study findings made sense from a biological perspective. “I wasn't surprised by the idea that brain amyloid-beta levels couldn't be neatly divided into positive or negative,” with values below a dividing line suggesting no worries.

“It creates a false expectation that if an amyloid level is negative, there is no risk. Cutoff points are useful, but they are inherently over-simplifying,” he said.

Dr. Knopman cautioned, however, that the study only showed an association between accumulating brain amyloid in the so-called negative range and subtle memory decline, not that the amyloid was causing the memory changes.

“I think the most plausible explanation is that brain amyloid elevations are a marker for neurodegeneration. The elevation of amyloid is a proxy for some other neurodegenerative process that, in the end, is causing neuronal dysfunction and death,” said Dr. Knopman.

Dr. Knopman said PET scan studies such as this latest one in Neurology are invaluable to Alzheimer's researchers because being able “to image the pathophysiology in real time in real people over the course of the disease, not to mention longitudinally, is fundamentally changing how we view the disease.”

Roberta Brinton, PhD, an Alzheimer's researcher who is professor in the departments of pharmacology and neurology at the University of Arizona, said the findings from the new study align with those of a study she conducted using PET scan and MRI that found that “amyloid-beta can accumulate very early in life,” perhaps well before any clinical signs. She is particularly interested in what happens during the hormonal transition that occurs as women approach menopause.

“We have evidence in the perimenopausal and menopausal female that amyloid-beta is already accumulating” when women are in their forties and fifties, said Dr. Brinton, whose findings were published last year in Neurology.

“At some point in this amyloid-beta negative phase, something is happening that leads to the development of the hallmark pathology of Alzheimer's,” Dr. Brinton said. “Our data indicate that decline in glucose metabolism in the brain is an early key driver activating hallmark pathologies of AD.”

While it would be impractical outside of a clinical trial for people to have a PET scan of their brain to check for amyloid buildup, Dr. Brinton said there are already “clinically deployable metrics,” that are routinely gathered during a doctor appointment that could be useful for predicting AD risk.

Dr. Brinton said she and colleagues published findings in 2016 suggesting that “it is possible, using standard clinical and laboratory analyses, to detect women who are showing signs of impending metabolic dysfunction. The earliest indicators of metabolic dysfunction were associated with cognitive decline, which is a risk for later life development of AD.”

Paul B. Rosenberg, MD, associate professor of psychiatry and behavioral science at Johns Hopkins School of Medicine, said the new study provides useful information for researchers and drug developers who are looking at the very earliest signs and risk factors for AD to help develop targets for intervention. Dr. Landau's team “turned the clock backwards and found that below the (amyloid positive) cutoff you can still be at some risk, you can still be accumulating risk,” he said.


DR. ROBERTA BRINTON: “We have evidence in the perimenopausal and menopausal female that amyloid-beta is already accumulating” when women are in their forties and fifties.

“The changes may be small and [may] not be enough to make a clinical difference,” said Dr. Rosenberg, but he said that the subset of seemingly vulnerable people in the negative range may be of interest to developers of anti-amyloid therapies because they may be the people most likely to benefit from such therapies.

Dr. Rosenberg is a site investigator for a large multicenter trial known as A4 (Anti-Amyloid in Asymptomatic Alzheimer's) that is testing an anti-amyloid antibody developed by Eli Lilly. The study involves 1,500 persons between the ages of 65 and 85 who have evidence on PET scan evidence of amyloid buildup but no clinical signs of cognitive decline. They are given monthly infusions of the drug for five years and being tracked with follow-up brain scans and cognitive tests.

DISCLOSURES: Dr. Landau reports consulting income from Genentech, Avid Radiopharmaceuticals, Janssen Pharmaceutical, Biogen, Cortexyme, and NeuroVision. The other investigators reported no disclosures relevant to the study.


• Landau SM, Horng A, Jagust WJ; for the Alzheimer's Disease Neuroimaging Initiative. Memory decline accompanies subthreshold amyloid accumulation Neurology 2018; Epub 2018 Mar 23.
    • Mosconi L, Berti V, Quinn C, et al. Sex differences in Alzheimer risk: Brain imaging of endocrine vs chronologic aging Neurology 2017: 89(13): 1382–1390.