β-Amyloid is Associated with Reduced Cognitive Function in Healthy Older Adults

A ccumulation of β-amyloid peptide in the brain parenchyma and vasculature is associated with Alzheimer’s disease. 1 Relative to healthy comparison subjects, plasma β-amyloid levels are elevated in dementia and in the sporadic Alzheimer’s disease risk factor e4 variant of apolipoprotein E (APOE). 2 However, the relationship between cognitive impairment and disease course is complex. Many researchers find no evidence for a link between β-amyloid and Alzheimer’s disease progression, 3 , 4 though this may reflect the nonlinear relationship between cognitive function and β-amyloid levels. For example, recent longitudinal evidence suggests that elevated plasma β-amyloid levels are present in some individuals before and in the earliest stages of the Alzheimer’s disease process but decrease over time. 5

The relationship between cognitive function and β-amyloid elevation in older adults without neurological conditions is currently unknown. Greater understanding of this relationship may accord key insights into the mechanisms by which β-amyloid affects the brain in the early stages of Alzheimer’s disease. Based on the current literature suggesting that elevated plasma β-amyloid levels are associated with aging and are present prior to the early stages of Alzheimer’s disease, we hypothesized that there would be a negative relationship between cognitive performance and plasma β-amyloid levels.

METHOD

The following procedures were approved by the local institutional review board. All participants provided written informed consent prior to study involvement.

Participants

Participants were recruited from local senior community centers and programs using fliers. Neither the centers nor the programs were providing any form of medical or psychiatric treatment to the participants. Participants met the following inclusion criteria: ages 60–85 years and proficiency with the English language. The exclusion criteria were current significant medical conditions (e.g., recent myocardial infarction, heart failure) or history of significant neurologic or psychiatric history (e.g., Alzheimer’s disease, stroke, schizophrenia). A total of 35 participants were eligible for current analyses using these criteria. See Table 1 for a summary of demographic and medical characteristics.

Enlarge table

Instrumentation

β-Amyloid 1–40 Detection

Serum β-amyloid 1–40 levels were determined using the Signet β-amyloid 1–40 ELISA assay kit (Covance Labs, Dedham, Mass.). This ELISA sandwiches β-amyloid 1–40 between antibodies that react with the amino terminal and carboxy terminal components of the β-amyloid 1–40 molecule. The manufacturer’s directions were followed, and specimens were tested in duplicate.

Neuropsychological Function

Participants completed a brief neuropsychological battery comprised of tests of global cognitive ability (Mini-Mental State Exam [MMSE]); 6 verbal memory (Hopkins Verbal Learning Test), 7 , 8 attention and psychomotor speed (Trail Making Test A, digit symbol coding), 9 , 10 working memory (Letter Number Sequencing), 10 executive function (Trail Making Test B; Frontal Assessment Battery), 9 , 11 and language (Boston Naming Test—Short Form; animal naming). 12 , 13 The average performance on all tests fell within expected ranges.

Self-Report Measures

In addition to the neuropsychological tests, participants were asked to report any history of medical or psychological conditions using a checklist format. Participants also reported subjective cognitive function using the cognitive difficulties scale. 14

Procedure

Participants provided written informed consent, underwent fasting blood draw, and were provided a small snack. Self-reported history of medical and psychological disorders was obtained from participants on a brief checklist. The neuropsychological test battery was then administered by trained research team members under the direction of a clinical neuropsychologist (JS, MBS). Specimens were collected by venipuncture in serum separator tubes and allowed to clot for 30 minutes at room temperature. After centrifugation for 15 minutes, the serum was aliquoted into 1.8 ml Nunc™ CryoTube vials and frozen at −80°C until the time of analysis. All variables showed expected distributions, and no transformations were performed.

RESULTS

Relationship between β-Amyloid and Demographic/Medical Variables

First, bivariate Pearson and Spearman correlations were conducted to determine the relationship between β-amyloid and demographic or medical variables, so that any significant variables could be utilized as covariates in subsequent analyses. No significant relationships between β-amyloid and any demographic or medical variable or cognitive difficulties scale total score emerged (see Table 2 ).

Enlarge table

Relationship between β-Amyloid and Neuropsychological Measures

Bivariate Pearson correlations between β-amyloid and neuropsychological measures were then conducted (see Table 3 ). Analyses showed significantly increased β-amyloid was associated with poorer global cognition (MMSE, r=−0.41, p<0.01), delayed verbal recall (Hopkins Verbal Learning Test delayed recall, r=−0.42, p<0.01), delayed verbal recognition (Hopkins Verbal Learning Test true hits, r=−0.34, p<0.05), semantic verbal fluency (animal naming, r=−0.32, p<0.05), and confrontation naming (Boston Naming Test—Short Form; r=−0.32, p<0.05). These values represent medium to large effect sizes.

Enlarge table

DISCUSSION

Consistent with predictions, the current study demonstrates an inverse relationship between β-amyloid and performance in several cognitive domains, including global cognition, verbal learning and delayed recall, semantic verbal fluency, and confrontation naming. These relationships exist in the presence of intact global cognition, as the average MMSE score in the sample was above 27.

Although the mechanisms for the relationship between cognitive function and β-amyloid are not entirely clear, the cognitive domains negatively associated with β-amyloid level in the current study reflect those abilities that are frequently impaired in the early stages of Alzheimer’s disease. 15 Such findings suggest that a percentage of the participants in the current study, despite no history of dementia and current global cognition within normal ranges, may convert to pre-Alzheimer’s disease (i.e., mild cognitive impairment) in the near future. This would be expected, in fact, based on the prevalence of Alzheimer’s disease in older adults living in the community 16 and would be consistent with longitudinal evidence demonstrating elevated β-amyloid in some individuals prior to early stages of the Alzheimer’s disease process. 5

A primary limitation of the current study includes its cross-sectional methodology, which does not permit examination of the relationship between β-amyloid and cognitive function over time. Given previous research indicating that β-amyloid level is elevated prior to developing Alzheimer’s disease, 5 and that base rates suggest that a proportion of the current study’s sample will develop Alzheimer’s disease in the future, longitudinal study is essential to the investigation of the relationship between β-amyloid and cognitive function in participants who develop Alzheimer’s disease in the future versus those who remain cognitively intact. Future studies would also benefit from the use of an expanded neuropsychological test battery (e.g., visual memory, specific aspects of attention and executive function) and a detailed assessment of family history of neurological conditions (e.g., Alzheimer’s disease, vascular dementia) to further clarify relationships. Similarly, studies employing large sample sizes are needed to help clarify the possible effects of different medications on both β-amyloid and cognitive function in healthy older adults. A growing number of medications have been suggested as having neuroprotective effects and may be implicated in β-amyloid-related processes.

Future research should also consider mediating variables between β-amyloid and cognition, particularly the possible moderating effects of various medical conditions. For example, recent research indicates that there are strong relationships between cardiovascular and physical health and both β-amyloid and cognition. 17 – 19 Such findings would suggest that improved cardiovascular fitness could suppress β-amyloid accumulation and thus influence its relation to cognitive function. Recent studies showing that exercise provides neuroprotection may be illustrating this possible effect. 20 Similarly, there is growing evidence that serum markers associated with cardiovascular dysfunction may be implicated in the development of Alzheimer’s disease. For example, a study by Irizarry et al. 21 links C-reactive protein to elevated risk of Alzheimer’s disease and suggests that homocysteine may interact with homocysteine to potentiate neurodegeneration in Alzheimer’s disease. Further work, particularly pathology studies, might determine the mechanisms by which β-amyloid is associated with cognitive function in healthy older adults and clarify their long-term outcome.

In summary, the current study extends previous work to show that cognitive function is inversely related to β-amyloid level in healthy older adults. Further work is needed to clarify possible mechanisms for these findings, particularly longitudinal and pathology studies that consider possible mediator variables.