A polygenic risk score (PRS) predicts cognitive decline in the APOE3 population in an early Alzheimer’s disease clinical trial cohort
Aparna Vasanthakumar, Qian Gao, Paula Daunt, Simon Flint, Alex Gibson, Hana Florian, Jeffrey Waring- Psychiatry and Mental health
- Cellular and Molecular Neuroscience
- Geriatrics and Gerontology
- Neurology (clinical)
- Developmental Neuroscience
- Health Policy
- Epidemiology
Abstract
Background
Alzheimer’s disease (AD) is one of the most common forms of dementia globally and is defined by accumulation of extracellular amyloid plaques and intracellular tau‐containing neurofibrillary tangles in neuronal cells. These pathological changes occur more than a decade prior to symptom onset. Given that disease modifying therapies are geared towards asymptomatic/early AD patients, identification of asymptomatic individuals at risk of progression is key. Genome‐Wide Association Studies(GWAS) for AD have implicated many single nucleotide polymorphisms (SNPs) associated with disease risk. Whereas a minority of early onset AD cases are driven by mutations in genes, sporadic AD cases have complex genetic architecture and are highly heritable. Polygenic Risk Scores (PRS), which aggregate genome‐wide risk loci, have been shown to predict AD onset with AUCs of 75‐84%.
Methods
The PRS approach used in this study is built from a sum of weighted contributions of over 100,000s SNPs associated with disease. We utilized PRS (genoSCORE™) to predict cognitive function decline in a cohort of 280 AD patients in a Ph2 clinical trial for Tilavonemab, a monoclonal antibody targeting the N‐terminus of human tau. Additionally, genoTOR™ was calculated to identify pathways most impacted by the genetic variants enriched in the cohort.
Results
Given that tilavonemab did not alter Clinical Dementia Rating‐Sum of Boxes(CDR‐SB), we tested differences in cognitive decline between high PRS and low PRS individuals defined using a threshold of 0.6 in the entire cohort and found no difference. Upon stratification by APOE status, higher PRS(> = 0.6) predicted an increase in CDR‐SB of an average of 1.0 at week 72 (p<0.1) and week 96 (p<0.05) compared to participants with lower PRS(<0.6) within the APOE3 population. Additional analyses of the genoTOR™ signaling pathways are ongoing.
Conclusions
The results of these analyses demonstrate that genoSCORE™ may be helpful to enrich for APOE3 individuals most susceptible to cognitive decline over time. Since the majority of genetic risk studies have focused on an APOE4 population, this method provides a novel tool to enrich for individuals at high risk for cognitive decline over time in a less studied population in which there is difficulty with prediction of disease progression.