Matthews DC, Andrews RD, Ordonez CE, Schmidt ME.
Methodologic considerations in the acquisition and analysis of amyloid imaging data for pharmaceutical clinical trials.
Human Amyloid Imaging 2011 Meeting Abstracts. 2011 Jan 15;
The availability of amyloid imaging agents for use with Positron Emission Tomography has provided a valuable
biomarker to measure amyloid burden as an indicator of Alzheimer's disease, and may be used to evaluate the
potential of anti-amyloid therapeutic candidates to modify disease progression. To date, much research has
focused upon the application of these agents to discriminate between diseased and normal subjects, and to
study the temporal progression of disease pathology as it relates to clinical progression. However, application
to drug development studies significantly raises the threshold for the degree of sensitivity, reproducibility, and
power required to predict clinical outcomes. Sources of variability that impact results and power must therefore be
understood, identified, and controlled to the extent possible. Our work has focused upon the analysis of the many
sources of variability in amyloid imaging data collected using [11C]PIB and [18F]AV45 and provided through ADNI,
and upon development of methods of quality control and analysis to optimize study power. Sources of variance
include image acquisition parameters, instrument and image reconstruction characteristics, subject artifact, image
pre-processing choices, and analysis methods. Specific factors include anatomical inclusion, noise at the edge
of the scanner, field of view, subject motion, image acquisition time window, image alignment, and selection of
reference and target regions of interest. To evaluate the potential contributions of these factors, we selected a set
of [11C]PIB dynamic scans. Scans were evaluated with regard to image quality and completeness, movement,
impact of different acquisition time windows, use of different ROI definitions, impact of reference region variance,
and use of voxel-based methods, and results were compared. The results of this survey identify the relative impact
of sources of amyloid image data variability and may lead to practical actions that can support the reliability of
imaging endpoints in both patient selection and drug evaluation.
* Data used in the preparation of this article were obtained from the Alzheimer's Disease Neuroimaging Initiative
(ADNI) database (www.loni.ucla.edu\ADNI). As such, the investigators within the ADNI contributed to the design
and implementation of ADNI and/or provided data but did not participate in analysis or writing of this report. ADNI
investigators include (complete listing available).