Genetic and Environmental Based Risk Model to Predict Cognitive Decline in a Healthy Population

Over the years, research has attempted to understand the etiology and pathophysiology of Alzheimer’s disease (AD), as well as the key risk factors involved in pathogenesis. While the underlying biology of AD is better understood due to these findings, clinical trials focused on AD treatment remain inconclusive, underscoring the importance of transitioning away from treatment and towards prevention in order to address the growing prevalence rates of AD. Thus, the overarching goal of this longitudinal study is to develop a risk model, utilizing six genetic and environmental based risk factors related to AD, to predict cognitive change using neuropsychological testing over a four-year period in a cognitively healthy population. While most studies to date have tested their predictive accuracy using clinical diagnoses, we explore the use of cognitive change as an endophenotype with high predictive accuracy, giving us more power with small sample sizes while also catching the first signs of disease onset, as cognitive deficit is a hallmark feature of AD. First, we explore the role of two individual genes in relation to cognitive change: apolipoprotein E (APOE), a well-recognized gene in the literature and a polymorphic, deoxythymidine tract in intron 6 of translocase of outer mitochondrial membrane (TOMM40’523), a highly debated gene in the field. Second, we explore the role of minor risk genes that have reached genome-wide significance using a polygenic risk score (PRS) that sums their effect. Lastly, because no studies to date have taken environmental risk factors into consideration, we explore the effect of three modifiable risk factors commonly associated with AD: systolic blood pressure, body mass index and a self-reported count of five metabolic conditions. We hypothesize that individuals with the highest genetic risk scores and the largest number of environmental risk factors will show the most cognitive decline over four years. Analyzing the role of these variables in four subdomains (APOE, PRS, TOMM40, metabolic health concerns), we present a method that provides insight for future studies limited by sample size and cross-sectional data, while also presenting novel information regarding the effect of TOMM40 and metabolic health concerns on AD risk. This approach may assist future clinical trials in screening for healthy, high-risk individuals that may benefit from early intervention. Lastly, our model could provide novel insights into the early pathogenesis of AD.