SUMMARY/ABSTRACT
This application is in response to PAR-17-054, which calls for combining multiple existing cohorts in order to
improve statistical power and clarify risk and protective factors for Alzheimer’s disease and related dementias
(AD/ADRD). While it is well-recognized that ADRD occurrence reflects the influences of multiple genes and
multiple environmental and lifestyle risk and protective factors, designs to elucidate potentially informative
gene-environment interplay have been rarer. Consequently we propose to employ a large consortium of
longitudinal twin studies of adult development and aging, with measures of risk and protective factors across
different life stages, to address key questions about etiological mechanisms in ADRD. By effectively treating
one twin as the control for the other, we will test for risk or protection associated with lifestyle, health, and
psychosocial factors while controlling for genetic factors, thereby strengthening causal inferences from
observational studies. We will also use twin designs to test the extent to which the association between risk or
protective factor and ADRD reflects shared genetic or shared environmental explanations. In addition to
within-pair and quantitative twin models, we will use polygenic risk scores (PRS) as indicators of individual
genetic risk for ADRD to test whether genetic risk for ADRD alters susceptibility to other risk and protective
factors, and PRS for specific risk and protective factors to test whether genetic risk for these factors alter their
association with ADRD. We especially focus on clarifying the nature of the relationship between education and
ADRD; midlife obesity, vascular risk, depression, and physical activity; and sex or gender differences in
genetic risk, exposure to specific risk factors, susceptibility to specific risk factors, and sex differences in
genetic interactions with specific risk factors. These questions importantly inform the design of interventions to
prevent or slow occurrence of dementia. The consortium on Interplay of Genes and Environments across
Multiple Studies (IGEMS) includes eight twin registries in the U.S., Sweden, Denmark, Finland, and Australia.
IGEMS brings nearly 50,000 individual twins, with over 5000 identical twin pairs for within-pair difference
models, nearly 7000 same-sex fraternal twin pairs, and over 3000 opposite sex pairs for sex difference
models. Outcomes include: clinically diagnosed dementia; dementia diagnoses obtained by linkage to national
health registries; psychometrically determined MCI; scores on a latent dementia indicator based on cognitive
and functional evaluations. A large subset of IGEMS participants has genome wide genotyping from which we
have computed PRS. For analyses of risk and protective factors, mediators, and covariates, we have data
from surveys collected in midlife as well as later and from linkages to administrative data, e.g., conscription
records and health registries. Already harmonized measures include: education, occupation, cognitive tests,
BMI, depression, anxiety.
