A global resilience score captures associations with AD pathology and predicts cognitive decline in cognitively healthy older adults

Abstract

AbstractBackgroundThe concepts of resilience and reserve were developed to help explain the variability in the relationship between pathology and cognition in older adults. We created a global resilience score that considers cognitive reserve, brain reserve, and cognitive resilience. We examined the relationship between global resilience and cross‐sectional PET measures of amyloid (PiB) and tau (FTP) and longitudinal change in cognition.Method130 cognitively normal older adults over the age of 60 (76.13±6.13 years, 76% F, 55% PiB+) were included in the analysis. Confirmatory factor analysis was used to create composite scores for episodic memory (EM) and executive function (EF). A partial least squares path model was used to derive a global resilience factor, a second‐order latent variable, from three first‐order latent variables estimating cognitive resilience, brain reserve, and cognitive reserve (Fig 1). PET data were coregistered to the structural MRIs and extracted from FreeSurfer‐segmented regions. A global PiB DVR and partial volume corrected FTP SUVR regional values were calculated. Longitudinal change in cognition was calculated as the slope of the linear fit over time for EM and EF.ResultThere were no significant relationships between global resilience and global PiB DVR (Fig 2A). Greater global resilience score was associated with lower entorhinal (ERC) FTP SUVR across the whole sample (R2 = 0.079, p = 0.001) and in PIB‐ (R2 = 0.062, p = 0.032) and PIB+ individuals (R2 = 0.088, p = 0.029) separately (Fig 2B). In PiB+ individuals, lower global resilience scores predicted greater decline in EM (R2 = 0.144, p = 0.008) and in EF (R2 = 0.098, p = 0.030) with PiB‐ individuals showing no significant relationships (Fig 3).ConclusionGlobal resilience score is associated with ERC tau and predicts cognitive change in amyloid+ individuals for both EM and EF. Our model reflects the proposed sequence of biomarker change in AD as global resilience is unassociated with PiB DVR but significantly correlated with ERC tau and cognitive change. These results indicate that a computed global resilience variable derived from measures of cognitive reserve, brain reserve and cognitive resilience is related to AD pathology and cognitive trajectory.