ARTICLE IN BRIEF
At the Alzheimer's Association International Conference, investigators reported data from separate trials indicating that subjective cognitive decline (SCD) correlates with an increase in amyloid burden, a higher likelihood of carrying Alzheimer's disease (AD) risk alleles, and an increased risk of conversion to AD or mild cognitive impairment within 10 years. Together, they strengthen the case that SCD may be the earliest predictor of more serious cognitive decline.
When a cognitively normal patient mentions that her memory is not quite what it used to be, is this the complaint of the worried well? Or the earliest sign of future Alzheimer's disease (AD)? New studies indicate that subjective cognitive decline (SCD) correlates with an increase in amyloid burden, a higher likelihood of carrying AD risk alleles, and an increased risk of conversion to AD or mild cognitive impairment within 10 years. Together, they strengthen the case that SCD may be the earliest predictor of more serious cognitive decline.
The new studies, presented at the Alzheimer's Association International Conference in Boston in July, add to the growing recognition that SCD is a clinically meaningful concept, with implications for understanding the preclinical stages of AD.
“People can have insight [about their cognitive condition] during the early part of the disease,” said David Bennett, MD, professor of neurology at Rush University Medical Center in Chicago. Dr. Bennett, who was not involved in the new studies, was among the earliest to study the phenomenon, and has led several longitudinal studies on memory and aging, including the National Institute on Aging-funded Religious Orders Study and the Rush Memory and Aging Project.
Rebecca Amariglio, PhD — a clinical neuropsychologist at Brigham and Women's Hospital in Boston, and an instructor in neurology at Harvard Medical School — studied the level of amyloid accumulation in the brains of 131 cognitively normal individuals who were part of the Harvard Aging Brain Study. Subjects were given a variety of questionnaires inquiring about memory concerns, and underwent PET imaging with Pittsburgh compound B to detect amyloid. “We found that the greater the concern for memory in everyday life, the greater the amyloid build-up in the brain,” Dr. Amariglio said. The relationship was true for those with either low or high cognitive reserve, as judged by education, occupation, and reading ability, and remained significant (at p=0.001) even after controlling for depression.
Executive function concerns were positively related to amyloid burden in the high, but not the low, cognitive reserve group. “Concerns may be a stronger indicator of amyloid build-up for the high education, high reading ability group,” she said. The reason for the discrepancy between the two groups is unclear, but it may be that someone with high reserves is engaged in more challenging everyday activities, “and if there's a dip they're more inclined to notice.”
There was no correlation between amyloid accumulation and informant reports of memory impairment, suggesting that, at the earliest stages of memory decline, a subject's own knowledge about their cognitive ability is superior to that of those around them.
VERBAL MEMORY DECLINE
Further evidence supporting an objective basis for SCD came from a study by Cecilia Samieri, PhD, of Research Center Inserm U897 in Bordeaux, France, and colleagues at the Nurses' Health Study, and Brigham and Women's Hospital. Dr. Samieri studied verbal memory decline over six years in over 4000 individuals, including 975 who carried the apolipoprotein E4 (APOE4) allele, which increases risk for AD. All participants were age 70 or older and were asked seven questions assessing subjective report of memory symptoms. Questions included whether the subject had trouble remembering a short list, remembering recent events, or following a group conversation. Among both carriers and non-carriers, those who reported memory symptoms at baseline declined more over time, reflected as a negative slope in memory ability over time. But for those carrying an APOE4 allele, “a single subjective memory symptom was sufficient to predict verbal memory change,” Dr. Samieri said, while in non-carriers, three or more symptoms were necessary to predict memory decline.
For APOE4 carriers reporting one memory symptom, the average slope in verbal memory ability over time was 0.05 (95% confidence interval 0.08, 0.01) standard units, whereas for non-carriers reporting one symptom, it was 0.01 (-0.02, 0.01), not significantly different from zero slope. The results confirm that self-report of memory concerns may be useful for identifying those at greater risk of memory decline, especially in APOE4 carriers, she said.
A FUTURE DIAGNOSIS
People who report a decline in memory over the course of a year are much more likely to be diagnosed with mild cognitive impairment (MCI) or AD in the future, compared with those who do not, according to a study from Richard Kryscio, PhD, professor of biostatistics at the College of Public Health at the University of Kentucky.
Dr. Kryscio's study was done as part of the Biologically Resilient Adults in Neurological Studies, a long-term assessment of cognitively normal aging individuals. The mean age at baseline of the 539 individuals included in the study was 73 years. Each was asked annually, “Have you noticed a change in your memory since your last visit?” More than half answered “yes” at some point during follow-up.
Dr. Kryscio and his colleagues found that those answering “yes” to that question were 2.8 times as likely to be diagnosed with MCI or AD during follow-up, compared with those answering “no.” “However,” Dr. Kryscio noted, “answering ‘yes’ is no guarantee that MCI or dementia will follow.” Indeed, only about half of those with memory complaints converted to MCI or AD during the study. One hundred sixty subjects converted to MCI or AD during follow-up, with transition occurring an average of nine years after baseline. Of those 160, 72.5 percent reported a change in memory before the transition, compared to 48.8 percent of the non-converters (p < 0.0001). For those who reported a memory change and later converted, the conversion occurred an average of six years after the first report.
In sum, Dr. Kryscio said, “Although not all older persons with subjective memory complaints proceed to a serious cognitive impairment, our data indicate that these self-reported concerns should be taken very seriously by clinicians.”
THE CLINICAL IMPACT
Ronald C. Petersen, MD, PhD, professor of neurology and director of the Alzheimer's Disease Research Center of the Mayo Clinic in Rochester, MN, who was not involved in any of these studies, concurred. “It is a trigger for the physician to drill down and ask the patient more about what kinds of changes they are experiencing.” Or in Dr. Amariglio's words, “It's a matter of just having your antenna up a little bit more.”
The most immediate impact of these studies may be on future clinical trials. The confirmation that SCD has important correlates in both genetics and pathology, and that it predicts increasing decline, could lead to use of SCD as an inclusion criterion for prevention trials, akin to those under way now, testing therapies in people without cognitive impairment but with inherited forms of AD or demonstrated amyloid burden. “These are the very people with whom we would want to enrich our clinical trials,” Dr. Amariglio said.