Regional tau pathology and loneliness in cognitively normal older adults

Abstract

Loneliness is a perception of social and emotional isolation that increases in prevalence among older adults during the eighth decade of life. Loneliness has been associated with higher brain amyloid-β deposition, a biologic marker of Alzheimer's disease, in cognitively normal older adults, suggesting a link with preclinical Alzheimer's disease pathophysiology. This study examined whether greater loneliness was associated with tau pathology, the other defining feature of Alzheimer's disease, in 117 cognitively normal older adults. Using flortaucipir positron emission tomography, we measured tau pathology in the entorhinal cortex, a region of initial accumulation in aging adults with or without elevated amyloid-β, and in the inferior temporal cortex, a region of early accumulation typically associated with elevated amyloid-β and memory impairment. Loneliness was measured by self-report using the 3-item UCLA-loneliness scale. We found that higher tau pathology in the right entorhinal cortex was associated with greater loneliness, controlling for age, sex, and apolipoprotein E ε4, the Alzheimer's disease genetic risk marker. This association remained significant after further adjustment for socioeconomic status, social network, depression and anxiety scores, and memory performance. There was no association of inferior temporal cortical or left entorhinal tau pathology with loneliness. Exploratory whole-brain surface maps supported these findings and identified additional clusters correlating loneliness and tau in the right fusiform gyrus. These results provide further support for loneliness as a socioemotional symptom in preclinical Alzheimer's disease.

Fig. 1. Multivariable regression was performed for UCLA-loneliness scores, the dependent variable (higher score indicates greater loneliness).

Plotted are the estimated marginal means for Flortaucipir (FTP) partial volume-controlled (PVC) signal uptake ratio (SUVr) binding for the right entorhinal cortex. Covariates included age, sex, and apolipoprotein E e4 carrier status. For right entorhinal FTP binding: β = 1.37, 95% CI [0.28, 2.15], Cohen’s f² = 0.254

Fig. 2. Shown above is the vertex-wise map correlating UCLA-loneliness scores with Flortaucipir signal uptake ratio, covaried for age, sex and apolipoprotein E e4 carrier status.

PET data were partial volume-corrected using the extended Müller-Gartner correction for surface analysis. Red colors indicate positive associations and reflect the one-sided t-statistic with threshold t > 2.36 (p = 0.01)