Premium
P1–265: Amyloid PET screening for enrollment into Alzheimer's disease clinical trials: An effective enrichment strategy in a phase 1b clinical trial
Author(s) -
Barakos Jerome,
Purcell Derk,
Klein Gregory,
Suhy Joyce,
Oh Joonmi,
Sevigny Jeff,
Chiao Ping
Publication year - 2013
Publication title -
alzheimer's and dementia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 6.713
H-Index - 118
eISSN - 1552-5279
pISSN - 1552-5260
DOI - 10.1016/j.jalz.2013.05.490
Subject(s) - kappa , clinical trial , medicine , cohen's kappa , nuclear medicine , pathology , statistics , mathematics , geometry
between 300 brain regions were generated for each participant. A MonteCarlo simulation was then implemented to estimate the detection power of a statistical test in the single-site vs multi-site setting. For each site and each simulation, the participants were randomly assigned to either a " patient " or a "control" group and a difference in connectivity between both groups was simulated at several specific connection. The connections were selected based on a real drug study on the effect of dopenezil in AD (Goveas et al JMRI 2011). Results: The amplitude of inter-site bias was about 3-fold smaller than the within-site standard deviation. The power analysis confirmed that a single-site studywith 200 subjects had only amarginally superior statistical power than an analysis pooling 8 sites for an equivalent number of subjects, see Fig.1. In both cases,a high sensitivity (>0.95) could be achieved for the effect size observed by (Goveas et al JMRI 2011). Conclusions: Our results showed that (1) the systematic variations in resting-state connectivity across scanners have a mild amplitude and that, (2) the statistical power of a multi-site resting-state fMRI study approaches the one of a single-site study, suggesting the feasibility of multi-site clinical trials in AD.