Social isolation is linked to classical risk factors of Alzheimer’s disease-related dementias

Abstract

Alzheimer's disease and related dementias is a major public health burden - compounding over upcoming years due to longevity. Recently, clinical evidence hinted at the experience of social isolation in expediting dementia onset. In 502,506 UK Biobank participants and 30,097 participants from the Canadian Longitudinal Study of Aging, we revisited traditional risk factors for developing dementia in the context of loneliness and lacking social support. Across these measures of subjective and objective social deprivation, we have identified strong links between individuals' social capital and various indicators of Alzheimer's disease and related dementias risk, which replicated across both population cohorts. The quality and quantity of daily social encounters had deep connections with key aetiopathological factors, which represent 1) personal habits and lifestyle factors, 2) physical health, 3) mental health, and 4) societal and external factors. Our population-scale assessment suggest that social lifestyle determinants are linked to most neurodegeneration risk factors, highlighting them promising targets for preventive clinical action.

APOE alleles are associated with sex-specific structural differences in brain regions affected in Alzheimer’s disease and related dementia

“…We computed the interactions between the subject-specific expressions of HC-DN co-variation modes (canonical variates) and each APOE genotype (encoded as binary variables, such that participants who do not carry a given genotype were zeroed out). In doing so, we obtained 6 new population-wide indices, 1 for each distinct APOE genotype that we correlated, using Spearman’s coefficient, with 63 curated ADRD risk factors (a phenotype collection used previously [ 34 ]). We then performed an agglomerative clustering analysis that consisted of a nested merging of correlation coefficients with similar variance until all observations merged in a single cluster.…”

“… To test for risk-anatomy links, we computed the Spearman’s correlations between the population-wide HC and DN co-variation patterns, multiplied by each of the 6 APOE genotypes and the 63 preselected Alzheimer’s disease risk factors [ 34 ]. We performed an agglomerative clustering analysis on these Spearman’s correlations, which consists of repeatedly merging Spearman’s correlations with similar variance until all observations are merged into a single cluster.…”

“…We adopted a targeted approach by focusing on a set of 63 risk factors (collection of phenotypes used previously [ 34 ]), including classical cardiovascular and demographic traits, as well as social richness indicators recently linked to ADRD in the UKB cohort. The first step of the clustering analysis consisted of multiplying the z-scored canonical variates by each of the 6 one-hot encoded APOE genotypes (i.e., ɛ2/2, ɛ2/3, ɛ3/3, ɛ2/4, ɛ3/4, and ɛ4/4) such that participants without a given genotype were zeroed out.…”

APOE ɛ2 vs APOE ɛ4 dosage shows sex-specific links to hippocampus-default network subregion co-variation

“…We adopted a targeted approach by focusing on a set of 63 risk factors (collection of phenotypes used previously [34]), including classical cardiovascular and demographic traits, as well as social richness indicators recently linked to ADRD in the UKB cohort. The first step of the clustering analysis consisted of multiplying the z-scored canonical variates by each of the six one- hot encoded APOE genotypes (i.e., ɛ2/2, ɛ2/3, ɛ3/3, ɛ2/4, ɛ3/4, and ɛ4/4) such that participants without a given genotype were zeroed out.…”

“…We computed the interactions between the subject-specific expressions of HC-DN covariation modes (canonical variates) and each APOE genotype (encoded as binary variables, such that participants who do not carry a given genotype were zeroed out). In doing so, we obtained six new population-wide indices, that is, one for each distinct APOE genotype that we correlated, using Spearman's coefficient, with 63 curated ADRD risk factors (a phenotype collection used previously [34]). We then performed an agglomerative clustering analysis which consisted in a nested merging of correlation coefficients with similar variance until all observations got merged in a single cluster.…”