There are many active research projects accessing and applying shared ADNI data. Use the search above to find specific research focuses on the active ADNI investigations. This information is requested annually as a requirement for data access.
| Principal Investigator | |
| Principal Investigator's Name: | Pawel Markiewicz |
| Institution: | University College London |
| Department: | Medical Physics and Biomedical Engineering |
| Country: | |
| Proposed Analysis: | To ADNI Data Publication and Sharing Committee: I am requesting access to the Alzheimer's Disease Neuroimaging Initiative (ADNI) data to support my research in Amyloid imaging to prevent Alzheimer’s Disease (AMYPAD) [1], a component of the European Prevention of Alzheimer’s Dementia (EPAD) project [2]. EPAD is a pre-competitive clinical research program including 35 partners from both private and academic sectors, sponsored by the European Commission and the European pharmaceutical industry (via EFPIA), and conducted under the auspices of the Innovative Medicines Initiative (IMI). The main objective of EPAD is to conduct adaptive, multi-arm Proof of Concept (PoC) studies which can contribute for early decision on the development of drug candidates intended to intercept the progression of AD. Within that context, AMYPAD will support EPAD by incorporating amyloid PET imaging into this setting. Amyloid PET will be therefore used to identify subjects for EPAD and PoC trials, and to study the onset, dynamics, and clinical relevance of brain β-amyloid deposition in the spectrum from normal ageing, through subjective cognitive decline, towards mild cognitive impairment, and ultimately dementia due to AD. One of the work packages of AMYPAD (Monitoring Treatment – quantifying patient-specific efficacy) is devoted to longitudinal quantitative amyloid PET as a biomarker for preclinical and prodromal AD and as a potential treatment assessment endpoint in PoC studies. Longitudinal quantitative amyloid PET has been previously used for as such in dementia studies, especially those regarding amyloid targeting monoclonal antibodies. Moreover, quantitative amyloid PET is currently being used in conjunction with image visual reads to identify and select subjects at risk for AD. However, if this is to be used effectively in early, at risk populations such as the ones considered for EPAD, the sensitivity of quantitative analysis for early detection of elevated levels of brain amyloid and change over time must be better understood [3]. First of all, factors independent of the amyloid signal per se can affect measurement accuracy, and those can include changes in blood flow over time and brain atrophy, both of which are characteristic of AD progression. In addition, other factors such as APOE4, cognitive reserve, and comorbid diseases may either enhance or mitigate amyloid deposition. In this context, the ADNI data with multiple imaging modalities offers an ideal sample in which to test and model the impact of such changes over time. This will be essential for the inclusion of the most appropriate subjects from AMYPAD to EPAD PoC trial studies, for the determination of the sample size required to demonstrate a treatment effect on amyloid deposition, and for the identification of key covariates for inclusion in statistical analyses. We intend to use the ADNI image dataset for investigation of different study designs and corresponding power calculations for estimating sample sizes needed for observing significant changes of chosen image biomarkers. Initially, we will perform cross-sectional analyses with event-based disease models which allow for the estimation of current disease stage and prediction of the rate of progression. These first findings will subsequently be compared with different longitudinal analyses accompanied by relevant disease models. In particular, we aim to investigate the detection of global and regional changes of amyloid deposition in (1) early and late amyloid accumulators, as well as (2) at different cognitive stages, while being able to account for cognitive reserve, APOE4, neurodegeneration and other relevant factors. For that purpose, we will perform linear longitudinal analyses with two measurement waves and extended it to non-linear models when more than three measurement waves are available. Next, we will investigate the possibility of using a-priori information obtained from rich longitudinal measurements (in case of non-linear longitudinal analysis) for modelling disease progression with only two-time point longitudinal measurements. For the analysis of static images, we will determine SUVr values while considering different reference regions. Moreover, the impact of partial volume correction (PVC) on the power analysis will also be investigated [4]. For those purposes, subject specific T1-weighted MR images will be parcellated into multiple regions of interest (ROI) using a multi-atlas segmentation propagation strategy [5]. The resulting ROIs will be then propagated to the native PET space using robust affine transformations [6]. Analyses and publications of results will focus on the needs of EPAD: sample size determination for PoC trials (contingent on the intended subject population and anticipated treatment effect on brain amyloid burden) and optimising methods for longitudinal quantitative amyloid imaging in PoC trials. We would plan to exchange information on the analysis and modelling of imaging and clinical data with AIBL investigators as well as consult and include them in publications. 1. AMYPAD: http://www.amypad.eu/ 2. EPAD: http://ep-ad.org/ 3. Schmidt, M.E., Chiao, P., Klein, G., Matthews, D., Thurfjell, L., Cole, P.E. et al, The influence of biological and technical factors on quantitative analysis of amyloid PET: Points to consider and recommendations for controlling variability in longitudinal data. Alzheimers Dement 2015; 11:1050–1068. 4. Erlandsson, K., Buvat, I., Pretorius, P.H., Thomas, B.A., Hutton, B.F., A review of partial volume correction techniques for emission tomography and their applications in neurology, cardiology and oncology. Physics in Medicine and Biology 2012; 57(21), R119 5. M. J. Cardoso et al., Geodesic Information Flows: Spatially-Variant Graphs and Their Application to Segmentation and Fusion, IEEE Transactions on Medical Imaging 2015; 34 (9): 1976-1988 6. Modat M, Cash DM, Daga P, Winston GP, Duncan JS, Ourselin S: Global image registration using a symmetric block-matching approach. Journal of Medical Imaging (Bellingham, Wash) 2014, 1(2):024003. |
| Additional Investigators |
