Accurately measuring resilience to preclinical Alzheimers disease (AD) pathology is essential to understanding an important source of variability in cognitive aging. In a cohort of cognitively normal older adults (n = 123, age 76.75 ± 6.15 yr), we built a multifactorial measure of resilience which moderated the effect of AD pathology on longitudinal cognitive change. Linear residuals-based measures of resilience, along with other proxy measures (education and vocabulary), were entered into a hierarchical partial least-squares path model defining a putative consolidated resilience latent factor (model goodness of fit = 0.77). In a set of validation analyses using linear mixed models predicting longitudinal cognitive change, there was a significant three-way interaction among consolidated resilience, tau and time on episodic memory change (P = 0.001) such that higher resilience blunted the effect of tau pathology on episodic memory decline. Interactions between consolidated resilience and amyloid pathology on non-memory cognition decline suggested that resilience moderates pathology-specific effects on different cognitive domains.

Approximately 30% of older adults exhibit the neuropathological features of Alzheimer's disease without signs of cognitive impairment. Yet, little is known about the genetic factors that allow these potentially resilient individuals to remain cognitively unimpaired in the face of substantial neuropathology. We performed a large, genome-wide association study (GWAS) of two previously validated metrics of cognitive resilience quantified using a latent variable modelling approach and representing better-than-predicted cognitive performance for a given level of neuropathology. Data were harmonized across 5108 participants from a clinical trial of Alzheimer's disease and three longitudinal cohort studies of cognitive ageing. All analyses were run across all participants and repeated restricting the sample to individuals with unimpaired cognition to identify variants at the earliest stages of disease. As expected, all resilience metrics were genetically correlated with cognitive performance and education attainment traits (P-values < 2.5 × 10-20), and we observed novel correlations with neuropsychiatric conditions (P-values < 7.9 × 10-4). Notably, neither resilience metric was genetically correlated with clinical Alzheimer's disease (P-values > 0.42) nor associated with APOE (P-values > 0.13). In single variant analyses, we observed a genome-wide significant locus among participants with unimpaired cognition on chromosome 18 upstream of ATP8B1 (index single nucleotide polymorphism rs2571244, minor allele frequency = 0.08, P = 2.3 × 10-8). The top variant at this locus (rs2571244) was significantly associated with methylation in prefrontal cortex tissue at multiple CpG sites, including one just upstream of ATPB81 (cg19596477; P = 2 × 10-13). Overall, this comprehensive genetic analysis of resilience implicates a putative role of vascular risk, metabolism, and mental health in protection from the cognitive consequences of neuropathology, while also providing evidence for a novel resilience gene along the bile acid metabolism pathway. Furthermore, the genetic architecture of resilience appears to be distinct from that of clinical Alzheimer's disease, suggesting that a shift in focus to molecular contributors to resilience may identify novel pathways for therapeutic targets.