O3‐01‐03: Genetic factors associated with the rate of cognitive decline in Alzheimer's disease

Alzheimer's disease (AD) is characterized by tremendous heterogeneity in the rate of cognitive decline after diagnosis without clear explanation. While several consortia have identified genes affecting the risk of AD, none have looked at rate of cognitive decline within cases of AD. We performed a genome-wide analysis using rate of cognitive decline among AD cases as the outcome. We analyzed data from 332 AD cases from the Alzheimer's Disease Neuroimaging Initiative (ADNI) that had been previously genotyped. We used GEE to estimate the coefficient associated with an interaction term between SNP genotype and time with AD to predict a Mini-Mental State Examination (MMSE) score modified using item response theory (IRT). To replicate our top findings, we analyzed longitudinal clinical and genotyping data available in an independent sample of 347 AD cases from the Religious Orders Study (ROS) and the Rush Memory and Aging Project (MAP), analyzing the average summary Z-score across the entire neuropsychological battery. IRT score analyses had advantages over standard MMSE scores, including a lower genomic inflation factor and a more normal trait distribution. Among our best results, we identified a variant in the MPP7 locus (rs1457177) that was strongly associated with rate of cognitive decline in ADNI AD subjects (P=2.2×10−7) and also showed evidence of replication in ROS/MAP (P=0.03). MPP7 is a membrane-associated guanylate kinase which are involved in the assembly of protein complexes at sites of cell-cell contact and are found in synapses, adherens junctions, and tight junctions. We additionally discovered associations between rate-of-decline and polymorphisms in several additional loci of interest, including the validated AD susceptibility gene, PICALM (rs41452444, PADNI=0.03) and other known candidates, such as SORCS1 (rs7093634, PADNI=0.02, PROS/MAP=0.01). Both PICALM and SORCS1 are implicated in vacuolar sorting and protein trafficking, a cell biological pathway with growing evidence for an important role in AD pathogenesis. These data suggest genetic factors predict rate-of-decline in AD, and that the risk may be mediated by novel pathways and also those already known to affect AD risk.