Resting-state functional magnetic resonance imaging in a randomized clinical trial for Alzheimer's disease

Abstract

Resting-state functional connectivity has been used to study Alzheimer's disease, revealing underlying differences between groups of individuals at different levels of cognitive impairment, or groupwise changes over time. Most of these studies have employed a single scanner and/or a single scan per subject. In this study, we used data collected from a substudy to two randomized clinical trials in patients with mild-to-moderate Alzheimer's Disease to investigate the robustness of functional connectivity in a multicenter, longitudinal setting. This substudy was not powered to detect drug treatment effects. With a basic imaging charter for inter-site acquisition standardization, we found no changes to functional connectivity over time in the placebo group, despite observable changes in clinical test scores and regional brain atrophy. We infer that functional connectivity, as implemented in our study, is not sensitive enough to detect disease progression in our patients. Using an 8-week interval between scans (in both treatment and placebo groups), we found that our test-retest reproducibility was comparable with other published results. We also found that the dominant source of measurement variability was within individuals over time, compared with variability between subjects and sites. Finally, we propose a simple method of visual inspection to identify scans that should be excluded from longitudinal analysis.

Keywords

• fMRI
• Resting state
• Functional connectivity
• Alzheimer's disease
• Randomized clinical trial
• Reproducibility