Abstract
OBJECTIVE:Cortical thinning, previously identified during prodromal stages of Alzheimer's disease (AD), is a "candidate" biomarker implemented in AD clinical therapy trials. We investigated the effect of donepezil treatment on cortical thickness in mild cognitively impaired subjects with the amnestic syndrome of the hippocampal type, a prodromal at-risk group for progression to AD dementia.METHODS:Data were from a longitudinal analysis of a community-based multicenter suspected prodromal AD cohort diagnosed by the Free and Cued Selective Reminding Test (81 donepezil vs 92 placebo) enrolled in a double-blind, randomized, placebo-controlled parallel group design using donepezil (10 mg/day). The study started in November 2006 and concluded in August 2010. All subjects underwent 2 brain structural magnetic resonance imaging (MRI) scans, at baseline and at the end of the trial. Structural MRI images had been processed using the automated pipeline for longitudinal segmentation and surface reconstruction implemented in FreeSurfer. The primary outcome measure of this post hoc study was the annualized percentage change (APC) of cortical thickness.RESULTS:The donepezil group exhibited reduced APC cortical thinning compared to placebo in the rostral anterior cingulate (right: P = .048; left: P = .032), the orbitofrontal (right: P = .012; left: P < .048), and the right inferior frontal (P = .022) cortices and in the right insula (P = .010). These results were not statistically significant after Bonferroni correction likely due to insufficient power for cortical thickness measurements in the study group powered for the predefined hippocampus outcome.CONCLUSIONS:Our findings support the hypothesis that cortical thickness is a reliable candidate surrogate outcome in early predementia AD trials. In addition, donepezil treatment may have an impact on cortical structure/morphology in areas innervated by the medial and lateral cholinergic pathways.