Predictors of cognitive response to cholinergic treatment in patients with Alzheimer's disease dementia


      Regional brain volumes related to memory and executive function, such as hippocampus and cholinergic basal forebrain, may be potential surrogate markers for cholinergic treatment outcome in Alzheimer’s disease (AD).


      We selected 131 cases from the ADNI1 database with psychometric follow up and cholinesterase inhibitor treatment over 0.5 to 3 years. We determined baseline volumes of left and right hippocampus as well as cholinergic basal forebrain nuclei. Using linear mixed effects models, we assessed the main effect of baseline volumes as well as their interaction with time, controlling for demographic variables. Endpoints were composite scores for memory and executive function, respectively, as previously defined [1,2].


      We found a significant main effect of basal forebrain volume on baseline executive function and memory, whereas hippocampus volume was significantly associated only with memory, but not with executive function. Volumes of basal forebrain were not significantly associated with the individual variation in decline of memory or executive function. In contrast, hippocampus volume reached a trend level effect on trajectories of memory dysfunction (p = 0.093), indicating larger rates of decline with higher hippocampus volumes. This effect was replicated in a voxel based analysis, where trajectories of memory decline were almost exclusively associated with hippocampus volumes of both sides (p < 0.001, uncorrected for multiple comparisons; Figure 1). Individual trajectories of decline in executive function did not show significant associations with regional grey matter volume in the voxel-based analysis.


      Reduced basal forebrain volume is associated with worse performance in both executive function and memory, but does not predict individual trajectories of cognitive decline in AD dementia during cholinesterase treatment. In contrast, reduced hippocampus volume is exclusively associated with worse memory performance, and with less decline of memory function during follow-up. Cholinergic treatment may overcome the predictive effect of basal forebrain volume on subsequent cognitive decline, whereas a relatively preserved hippocampus volume predicts subsequent memory decline, possibly related to a floor effect in individuals with advanced hippocampus atrophy. References: 1. Gibbons et a. Brain Imaging Behav. 2012;6:517-27. 2. Crane et al. Brain Imaging Behav. 2012;6:502-16.
      Figure thumbnail fx1
      Figure 1Voxel-based analysis of the effect of individual trajectories of memory decline on grey matter volume.