Cerebral accumulation of amyloid beta (Abeta) protein is a pathological hallmark of Alzheimer disease (AD), and likely plays a fundamental role in its pathogenesis. Research has focused on Aa manipulation as a mechanism for reducing or treating AD, but the clinical utility of findings is limited by lack of a simple way to identify early those at high risk of AD; Abeta levels in the CSF have been measured in some settings, but are impractical for widespread clinical applications. Abeta appears in circulating plasma, and a central aim of this proposal is to investigate the value of plasma Abeta as an early indication of AD risk by exploring its relation to cognitive decline, an early stage in the path of AD development. An additional aim of this proposal is to explore the function of insulin in cognitive decline, and its interaction with Abeta; irecent research has established that insulin modulates levels of Abeta in vitro and in vivo, however, little work has examined the long-term effects of insulin on cognition, or the possibility of different effects in subgroups of the population. This proposal represents a collaboration between basic scientists and epidemologists to test novel hypotheses in men and women regarding plasma Abeta levels, insulin levels, and insulin secretion (as measured by c-peptide) in non-diabetics and their ability to predict development of cognitive decline. The investigation will be conducted within the Nurses' Health Study, including women currently aged 70- 81 years and followed since 1976, and the Physician's Health Study II including men aged 65 years and older and followed since 1982. From each cohort, we will select 2,000 subjects with stored plasma samples collected at two points over 10 years, and with serial tests of cognitive function (3 repeated interviews by telephone, at 1.5-year intervals). The test battery measures general cognitive status, verbal memory, executive function, speed of processing, and working memory; we have conducted extensive demonstrations of the validity and reliability of the telephone method compared to in-person assessments, and have previously detected strong relations with numerous established risk factors for decline. Thus, these two established populations provide a highly cost-efficient setting in which to examine biomarkers of early cognitive decline, and gender and genetic interactions.
Showing the most recent 10 out of 13 publications
