Blood Oxygen Level Dependant (BOLD) fMRI was performed at Screening (Baseline) and on Days 2 and 90. During each of these sessions, BOLD fMRI images were acquired in rapid succession as a flashing radial black and white checkerboard was presented alternately with a gray screen. This well established visual stimulus is known to produce a reliable increase in BOLD fMRI signal within the visual cortex region of the occipital lobe. The time course of the BOLD fMRI signal was used to assess the vascular reactivity. Imaging sites also acquired cerebral blood flow data using Arterial Spin Labeled (ASL) scans at Screening and on Days 2 and 90. A standard T1-weighted image was also acquired to aid image analysis. All efficacy scans were analyzed centrally. Geometric means are presented in the original scale and standard errors (SE) are presented in logarithmic (log e) scale.

BOLD fMRI was performed at Screening (Baseline) and on Days 2 and 90. During each of these sessions, BOLD fMRI images were acquired in rapid succession as a flashing radial black and white checkerboard was presented alternately with a gray screen. This well established visual stimulus is known to produce a reliable increase in BOLD fMRI signal within the visual cortex region of the occipital lobe. The time course of the BOLD fMRI signal was used to assess the vascular reactivity. Imaging sites also acquired cerebral blood flow data using ASL scans at Screening and on Days 2 and 90. A standard T1-weighted image was also acquired to aid image analysis. All efficacy scans were analyzed centrally. Geometric means are presented in the original scale and SEs are presented in log e scale.